CN106445405B - Data access method and device for flash memory storage - Google Patents

Abstract

A data access method and device for flash memory storage are provided. A data access method for a storage system, the storage system comprising a plurality of storage devices, the storage system providing a plurality of storage objects, a storage object being comprised of storage resources on a storage device, the plurality of storage objects comprising one or more writable storage objects and a plurality of read-only storage objects; the method comprises the following steps: responding to a write request, and writing data to the writable storage object in an additional write mode; if the writable storage object is full, setting the writable storage object as a read-only storage object; in response to a read request, data is read from the read-only memory object.

Description

A data access method and device for flash memory storage

technical field

The present invention relates to a high-performance storage system, and more particularly, the present invention relates to a data access method and apparatus for a storage system that separates read and write operations at the storage object layer during data output to a record carrier in the storage system.

Background technique

In the prior art, a solid state drive (SSD, Solid State Drive) is usually accessed by means of an LBA (Logical Block Address, logical block address). The data access generated in the file system needs to determine the LBA, and then access the corresponding SSD according to the LBA. This way of using LBA for data layout is more intuitive, and can improve the performance of data access through data caching and making full use of the locality of application data.

The prior art is evolved from the disk technology, which makes full use of the characteristics of the disk storage device with strong sequential access capability and weak random access capability. However, the way of accessing SSD through LBA cannot meet the requirement of storage performance consistency, and when the storage system is used for a period of time, the overall performance of the system will decrease due to a large number of data recovery operations.

In order to improve the write operation performance of the storage system, a cache is generally provided in the system for caching the written data. One way to provide a write cache is to create a cache in memory. After a write request to a file or block device is written to the cache in memory, the write operation can be returned to the host, and then the data can be asynchronously written to disk through a background operation. . This method is often called write-back. Another method is to wait for the data in the buffer to be synchronized to the disk before returning the completion of the write operation to the host, which is called write-through.

NVDIMM is a memory used for computing equipment. It has both the high-speed data access capability similar to DRAM (Dynamic Random Access Memory) and the data retention capability of non-volatile memory. Even if an unexpected power failure occurs, the data stored in the NVDIMM will not be lost. .

Referring to FIG. 1 , an iSCSI storage server using NVDIMM as a cache is disclosed in Chinese Patent Application Publication No. CN104239226A. When the storage server is working, all iSCSI read and write commands complete the corresponding read and write operations through the disk cache. When the storage server receives an iSCSI read command, it searches the disk cache first, and if it finds the corresponding data, it returns it directly to the client; if it does not find the corresponding data, it reads the data from the disk into the disk cache, and then returns To the client; when the storage server receives the iSCSI write command, it directly copies the data to the corresponding area of the disk cache.

In the prior art, the NVDIMM is used as both a read cache and a write cache, which places a high demand on the capacity of the NVDIMM. In the face of streaming data access requests, it is difficult to obtain high utilization in the existing caching methods, and frequent cache misses will cause the overall performance of the storage system to thrash.

SUMMARY OF THE INVENTION

One purpose of the present invention is to solve the problem of improving the writing performance on the premise that the data is not lost when the power is turned off, so as to realize the balance between the writing performance and the data reliability. Another object of the present invention is to provide an effective cache replacement mechanism for streaming big data, improve the utilization efficiency of the cache, reduce the performance bumps of the storage system, and enable the storage system to achieve delay consistency. Another object of the present invention is to propose a way of distributing data on a flash memory storage medium, so as to reduce the mutual influence between read and write operations and control the read operation delay within a certain range. Achieve consistent performance while enabling flash storage systems to achieve optimal performance. In addition, the purpose of the present invention is to enhance the service life of the flash memory while improving the system performance.

According to a first aspect of the present invention, there is provided a first NVDIMM-based data write caching method according to the first aspect of the present invention, comprising: receiving a first data write request, the first data write request indicating that the first data Writing a first address; Affecting receiving the first data write request, writing the first data into the NVDIMM; in response to the completion of the operation of writing the first data into the NVDIMM, sending an indication of the first data Write request completion message; in response to receiving the first data write request, also write the first data to the storage device; in response to the completion of the operation of writing the first data to the storage device, release the storage space occupied by the first data in the NVDIMM.

According to the first NVDIMM-based data write caching method according to the first aspect of the present invention, there is provided the second method according to the first aspect of the present invention, wherein the operation of writing the first data to the NVDIMM is the same as the writing of the first data. operations to the storage device are performed in parallel.

The first or second NVDIMM-based data write caching method according to the first aspect of the present invention provides a third method according to the first aspect of the present invention, wherein the writing the first data to the NVDIMM comprises: generating a first A data block, the first data block includes the first data, the first address and a sequence number, wherein each time a sequence number is generated, the sequence number is incremented; the first data block is written into all NVDIMMs described above.

The first to third NVDIMM-based data write caching methods according to the first aspect of the present invention provide a fourth method according to the first aspect of the present invention, further comprising: receiving a second data write request, the second data write The write request instructs to write the second data to the second address; in response to receiving the second data write request, a second data block is generated, and the second data block includes the second data, the second data address and sequence number; write the second data block to the NVDIMM.

According to the first to third NVDIMM-based data write caching methods according to the first aspect of the present invention, there is provided a fifth method according to the first aspect of the present invention, further comprising: in response to receiving a message of normal shutdown, sending a message to the NVDIMM Write the first mark.

According to the fourth NVDIMM-based data write caching method according to the first aspect of the present invention, there is provided the sixth method according to the first aspect of the present invention, further comprising: in response to receiving a message of normal shutdown, writing the first method to the NVDIMM a mark.

The sixth NVDIMM-based data write caching method according to the first aspect of the present invention provides the seventh method according to the first aspect of the present invention, further comprising: in response to power-on, if the first method cannot be read from the NVDIMM mark, the first data block and the second data block are read from the NVDIMM, and the first data block and the second data block The first data and the second data in the first data block and the second data block are written into the storage device.

The fourth NVDIMM-based data write caching method according to the first aspect of the present invention provides the eighth method according to the first aspect of the present invention, further comprising: in response to receiving a message of abnormal shutdown, writing the first method to the NVDIMM Two flags; in response to power-on, if the second flag is read from the NVDIMM, the first data block and the second data block are read from the NVDIMM, according to the first data block The first data and the second data in the first data block and the second data block are written into the storage device in ascending order of the sequence numbers in the second data block.

According to the aforementioned NVDIMM-based data write caching method according to the first aspect of the present invention, there is provided a ninth method according to the first aspect of the present invention, wherein in response to writing the first data to the NVDIMM, the first data is stored in the The storage space occupied on the NVDIMM is marked as occupied; in response to releasing the storage space occupied by the first data in the NVDIMM, the storage space occupied by the first data on the NVDIMM is marked as free.

According to the ninth NVDIMM-based data write cache method according to the first aspect of the present invention, there is provided the tenth method according to the first aspect of the present invention, wherein the first data is written when the first data is written into the NVDIMM into the memory space of the NVDIMM that is marked as free.

The aforementioned NVDIMM-based data write caching method according to the first aspect of the present invention provides an eleventh method according to the first aspect of the present invention, further comprising: writing the first address into the storage device.

According to a second aspect of the present invention, there is provided a first NVDIMM-based data write cache device according to the second aspect of the present invention, comprising: a receiving module configured to receive a first data write request, the first data write request Instruct to write the first data to the first address; the NVDIMM writing module is used to write the first data into the NVDIMM in response to receiving the first data write request; the message sending module is used to respond to the writing of the first data. After the operation of writing the first data into the NVDIMM is completed, a message indicating the completion of the first data writing request is sent; the storage device writing module is configured to, in response to receiving the first data writing request, send the writing first data to a storage device; and an NVDIMM release module, configured to release the storage space occupied by the first data in the NVDIMM in response to the completion of the operation of writing the first data to the storage device.

According to a third aspect of the present invention, there is provided a first NVDIMM-based data write caching method according to the third aspect of the present invention, comprising: receiving a first data write request, the first data write request indicating that the first data writing the first address; in response to receiving the first data write request, writing the first data into the NVDIMM; in response to the completion of the operation of writing the first data into the NVDIMM, sending an indication of the first data write request completion message; receive a second data write request, the second data write request instructing to write the second data to the second address; in response to receiving the second data write request, write the second data write request Two data are written into the NVDIMM; in response to the completion of the operation of writing the second data into the NVDIMM, a message indicating the completion of the second data write request is sent; a first storage data block is generated, and the first storage data block is including the first data, the first address, the second data, and the second address; writing the first block of storage data to a storage device; in response to writing the first storage data to storage After the operation of the device is completed, the storage space occupied by the first data and the second data in the NVDIMM is released.

The first NVDIMM-based data write caching method according to the third aspect of the present invention provides the second method according to the third aspect of the present invention, wherein the writing the first data to the NVDIMM includes: generating a first data block, The first data block includes the first data, the first address, and a sequence number; writing the first data block to the NVDIMM; and wherein the writing the second data to the NVDIMM includes : generate a second data block, the second data block includes the second data, the second address and the sequence number; write the second data block into the NVDIMM; and where the sequence number is generated each time , increment the sequence number.

According to the first or second NVDIMM-based data write caching method according to the third aspect of the present invention, there is provided the third method according to the third aspect of the present invention, further comprising: in response to receiving a message of normal shutdown, sending a message to the NVDIMM Write the first mark.

A third NVDIMM-based data write caching method according to the third aspect of the present invention provides a fourth method according to the third aspect of the present invention, further comprising: in response to power-on, if the first method cannot be read from the NVDIMM mark, the first data block and the second data block are read from the NVDIMM, and the first data block and the second data block The first data and the second data in the first data block and the second data block are written into the storage device.

According to the aforementioned NVDIMM-based data write caching method according to the third aspect of the present invention, there is provided a fifth method according to the third aspect of the present invention, wherein in response to writing the first data to the NVDIMM, the first data is stored in the The storage space occupied on the NVDIMM is marked as occupied; in response to releasing the storage space occupied by the first data in the NVDIMM, the storage space occupied by the first data on the NVDIMM is marked as free.

According to a fifth NVDIMM-based data write caching method according to the third aspect of the present invention, there is provided a sixth method according to the third aspect of the present invention, wherein the first data is written when the first data is written into the NVDIMM into the memory space of the NVDIMM that is marked as free.

According to a fourth aspect of the present invention, there is provided a first NVDIMM-based data write buffer device according to the fourth aspect of the present invention, comprising: a first receiving module configured to receive a first data write request, the first data write The write request instructs to write the first data to the first address; the first NVDIMM writing module is configured to write the first data into the NVDIMM in response to receiving the first data write request; the first message sending module, In response to the completion of the operation of writing the first data to the NVDIMM, sending a message indicating that the first data write request is completed; a second receiving module, configured to receive a second data write request, the second The data write request instructs to write the second data into the second address; the second NVDIMM write module is used to write the second data into the NVDIMM in response to receiving the second data write request; the second message is sent a module, configured to send a message indicating that the second data write request is completed in response to the completion of the operation of writing the second data to the NVDIMM; a data aggregation module, configured to generate a first storage data block, the first The storage data block includes the first data, the first address, the second data and the second address; a storage device writing module is configured to write the first storage data block into a storage device; and an NVDIMM release module, configured to release the storage space occupied by the first data and the second data in the NVDIMM in response to the completion of the operation of writing the first stored data to the storage device.

According to a fifth aspect of the present invention, there is provided a computer program which, when loaded into and executed on a computer system, causes the computer system to execute the first and third aspects of the present invention And one of the many methods offered.

According to a sixth aspect of the present invention, there is provided a computer comprising: a machine-readable memory for storing program instructions; one or more processors for executing the program instructions stored in the memory; the program The instructions are for causing the one or more processors to perform one of the methods provided in accordance with the first and third aspects of the present invention.

According to a seventh aspect of the present invention, there is provided a program that causes a computer to perform one of the methods provided according to the first and third aspects of the present invention.

According to an eighth aspect of the present invention, there is provided a computer-readable storage medium having a program recorded thereon, wherein the program causes a computer to perform the various methods provided according to the first and third aspects of the present invention one.

According to a ninth aspect of the present invention, there is provided the first write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein the write cache includes one or more free storage spaces, and provides a data structure for to index the one or more free storage spaces, the method includes: in response to receiving a request to release the first storage space of the write cache, looking up whether the storage space adjacent to the first storage space is free storage space; if the first free storage space connected to the first storage space is found, modify the first node in the data structure that indexes the first free storage space, so that the first node indexes the first free storage space. a free storage space and the first storage space; and if the first storage space connected to the first storage space is not found, a new node is added in the data structure for indexing the first storage space.

According to a ninth aspect of the present invention, there is provided a second write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein the write cache includes one or more free storage spaces, and a data structure is provided for indexing the one or more free storage spaces, the data structure includes a plurality of nodes, each node is used for indexing a free storage space, the method includes: in response to receiving the first storage space that releases the write cache request, find out whether the free storage space before and after the first storage space is connected to the first storage space;

If the first free storage space connected to the first storage space is found, the first free storage space and the first storage space are merged; and if the first free storage space connected to the first storage space is not found A space storage space, and a new node is added in the data structure for indexing the first storage space.

According to the second write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided a third write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein the combining said The first free storage space and the first storage space include: modifying a first node in the data structure that indexes the first free storage space, so that the first node indexes the first free storage space and the first free storage space. first storage space.

According to the second write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided a fourth write cache release method for streaming data processing according to the ninth aspect of the present invention. If the first free storage space and the second free storage space are adjacent to the first storage space, the first storage space and the second free storage space are merged, wherein the address of the first free storage space is in the first storage space. Before the storage space, and the address of the second space storage space is after the first storage space.

According to the aforementioned write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided a fifth write cache release method for streaming data processing according to the ninth aspect of the present invention, further comprising: in the first A flag is set in a storage space to indicate that the first storage space is free.

According to the aforementioned write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided a sixth write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein a pointer is provided, the pointer The pointed node indexes the first free storage space; in response to receiving a request for allocating storage space, the free storage space is searched from the node pointed to by the pointer.

The sixth write cache release method for streaming data processing according to the ninth aspect of the present invention provides the seventh write cache release method for streaming data processing according to the ninth aspect of the present invention, further comprising: if the The node pointed to by the pointer can satisfy the request for allocating storage space, and free storage space is allocated from the free storage space indexed by the node pointed to by the pointer in response to the request for allocating storage space.

According to the seventh write cache release method for streaming data processing according to the ninth aspect of the present invention, the eighth write cache release method for streaming data processing according to the ninth aspect of the present invention is provided, further comprising: if not found The third node that indexes the free storage space that can satisfy the request, and waits for the free storage space that can satisfy the request to appear.

According to the second write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided the ninth write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein the combining said The first free storage space and the first storage space include: adding a second node in the data structure for indexing the first storage space and the first storage space; and deleting the index in the data structure. A first node of free storage space.

According to the fourth write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided a tenth write cache release method for streaming data processing according to the ninth aspect of the present invention, including: modifying the data In the structure, the first node that indexes the first free storage space or the second node that indexes the second free storage space, so that the first node or the second node indexes the first free storage space, all the the first storage space and the second storage space.

According to the fourth write cache release method for streaming data processing according to the ninth aspect of the present invention, an eleventh write cache release method for streaming data processing according to the ninth aspect of the present invention is provided, including: in the data structure Add a second node to index the first free storage space, the first storage space and the second storage space; and delete the first node indexing the first free storage space in the data structure, and the index of the first node. the second node of the second free storage space.

According to the aforementioned write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided a twelfth write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein the write cache The storage space is organized as a ring buffer.

According to the aforementioned write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided the thirteenth write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein the node index is free The storage space includes the first address of the free storage space and the length of the free storage space referenced by the mark in the node.

According to the aforementioned write cache release method for streaming data processing according to the ninth aspect of the present invention, there is provided the fourteenth write cache release method for streaming data processing according to the ninth aspect of the present invention, wherein the plurality of nodes Sort by the address of the free storage space it indexes.

According to a tenth aspect of the present invention, there is provided the first write cache allocation method for streaming data processing according to the tenth aspect of the present invention, wherein the write cache includes one or more free storage spaces, and provides a data structure for to index the one or more free storage spaces, the data structure includes a plurality of nodes, each node is used to index a free storage space, and a first pointer is provided to point to one of the plurality of nodes, and the method includes : in response to receiving the first request for allocating the first storage space, search for free storage space from the first node pointed to by the pointer; if the first node can satisfy the first request, search for free storage space from the first node indexed by the first node The low address of the free storage space begins to allocate free storage space to respond to the first request; the first node is modified so that the first node index responds to the free storage space after the first request.

The first write buffer allocation method for streaming data processing according to the tenth aspect of the present invention provides the second write buffer allocation method for streaming data processing according to the tenth aspect of the present invention, further comprising: if so The first node does not satisfy the first request, and waits for the first node to have free storage space that can satisfy the first request.

The first write buffer allocation method for streaming data processing according to the tenth aspect of the present invention provides a third write buffer allocation method for streaming data processing according to the tenth aspect of the present invention, further comprising: if so If the first node does not satisfy the first request, traverse the plurality of nodes in sequence to search for free storage space that can satisfy the first request.

The third write cache allocation method for streaming data processing according to the tenth aspect of the present invention provides the fourth write cache allocation method for streaming data processing according to the tenth aspect of the present invention, further comprising: if not Find the node that indexes the free storage space that can satisfy the first request, and wait for the free storage space that can satisfy the first request to appear.

The aforementioned write buffer allocation method for streaming data processing according to the tenth aspect of the present invention provides a fifth write buffer allocation method for streaming data processing according to the tenth aspect of the present invention, further comprising: in response to receiving To the request to release the second storage space, find out whether the free storage space before and after the second storage space is adjacent to the second storage space; if the first free storage space adjacent to the second storage space is found, Then combine the first free storage space and the second storage space, and modify the node that has indexed the first free storage space to make the index of the combined first free storage space and the second storage space ; and if the first free storage space adjacent to the second storage space is not found, adding a new node in the data structure for indexing the second storage space.

The aforementioned write cache allocation method for streaming data processing according to the tenth aspect of the present invention provides a sixth write cache allocation method for streaming data processing according to the tenth aspect of the present invention, wherein nodes index free storage space It includes the first address of the free storage space and the length of the free storage space referenced by the mark in the node.

The aforementioned write buffer allocation method for streaming data processing according to the tenth aspect of the present invention provides a seventh write buffer allocation method for streaming data processing according to the tenth aspect of the present invention, wherein the plurality of nodes Sort by the address (in increasing order) of the free storage space it indexes.

The aforementioned write cache allocation method for streaming data processing according to the tenth aspect of the present invention provides an eighth write cache allocation method for streaming data processing according to the tenth aspect of the present invention, wherein the write cache The storage space is organized as a ring buffer.

According to an eleventh aspect of the present invention, an NVDIMM-based data write cache method is provided, comprising: receiving a first data write request, the first data write request instructing to write the first data to a first address; affecting upon receiving the first data write request, write the first data into the NVDIMM, wherein the NVDIMM includes one or more free storage spaces, and provide a data structure for indexing the one or more free storage spaces , the data structure includes a plurality of nodes, each node is used to index a free storage space, provides a first pointer, and points to one of the plurality of nodes; starts from the first node pointed to by the pointer to search for the free storage space; If the first node can satisfy the first request, allocate the first free storage space from the low address of the free storage space indexed by the first node; modify the first node so that the first node index responds the free storage space after the first request is received; write the first data into the first free storage space; in response to the completion of the operation of writing the first data into the NVDIMM, sending an indication of the first data Write request completion message; in response to receiving the first data write request, also write the first data to the storage device; in response to the completion of the operation of writing the first data to the storage device, release the storage space occupied by the first data in the NVDIMM.

According to a twelfth aspect of the present invention, a write cache release device is provided, wherein the write cache includes one or more free storage spaces, a data structure is provided for indexing the one or more free storage spaces, and the The device includes: a free storage space search module for, in response to receiving a request for releasing the first storage space of the write cache, to find out whether the storage space connected to the first storage space is free storage space; the storage space is merged A module, configured to modify the first node indexing the first free storage space in the data structure if the first free storage space connected to the first storage space is found, so that the first node indexes the The first free storage space and the first storage space; and an index adding module for adding a new node in the data structure for indexing if the first space storage space connected to the first storage space is not found the first storage space.

According to a thirteenth aspect of the present invention, a write cache allocation device is provided, wherein the write cache includes one or more free storage spaces, a data structure is provided for indexing the one or more free storage spaces, and the The data structure includes a plurality of nodes, each node is used for indexing a free storage space, and a first pointer is provided to point to one of the plurality of nodes, and the apparatus includes: a free storage space search module for responding to receiving the allocation For the first request for the first storage space, search for free storage space from the first node pointed to by the pointer; the storage space allocation module is configured to, if the first node can satisfy the first request, search for free storage space from the first node indexed by the first node. The low address of the free storage space starts to allocate free storage space to respond to the first request; the node modification module is used to modify the first node, so that the first node index responds to the idle after the first request. storage.

According to a fourteenth aspect of the present invention, there is provided an NVDIMM-based device for writing data, comprising: a receiving module configured to receive a first data writing request, where the first data writing request instructs to write the first data into the first address; the NVDIMM writing module is configured to, in response to receiving the first data writing request, write the first data into the NVDIMM, wherein the NVDIMM includes one or more free storage spaces for providing data The structure is used to index the one or more free storage spaces, the data structure includes a plurality of nodes, each node is used to index a free storage space, and a first pointer is provided to point to one of the plurality of nodes; the The NVDIMM writing module includes: a free storage space search module, configured to search for free storage space from the first node pointed to by the pointer; a storage space allocation module, configured to, if the first node can satisfy the first request, Allocate the first free storage space from the free storage space indexed by the first node; a node modification module, configured to modify the first node, so that the first node index responds to the free storage space after the first request; and a data writing module for writing the first data into the first free storage space; the NVDIMM-based device for writing data further includes: a message sending module for writing the first data in response to After the operation of writing data to the NVDIMM is completed, a message indicating the completion of the first data writing request is sent; the storage device writing module is configured to, in response to receiving the first data writing request, write the first data Writing to a storage device; an NVDIMM release module, configured to release the storage space occupied by the first data in the NVDIMM in response to the completion of the operation of writing the first data to the storage device.

According to a fifteenth aspect of the present invention, there is provided a computer comprising: a machine-readable memory for storing program instructions; one or more processors for executing the program instructions stored in the memory; the program The instructions are for causing the one or more processors to perform one of the methods provided in accordance with the ninth to eleventh aspects of the present invention.

According to a sixteenth aspect of the present invention, there is provided a program that causes a computer to execute one of the methods provided according to the ninth and eleventh aspects of the present invention.

According to a seventeenth aspect of the present invention, there is provided a computer-readable storage medium having a program recorded thereon, wherein the program causes a computer to execute the multiple functions provided according to the ninth and eleventh aspects of the present invention. one of the methods.

According to an eighteenth aspect of the present invention, there is provided a first data access method for a storage system according to the eighteenth aspect of the present invention, the storage system comprising a plurality of storage devices, the storage system providing a plurality of storage objects, The storage object is composed of storage resources on the storage device, and the multiple storage objects include one or more writable storage objects and multiple read-only storage objects; the method includes: responding to a write request, sending a request to the writable storage object The object writes data in an append/sequential write mode; if the writable storage object is full, the writable storage object is set as a read-only storage object; in response to a read request, from the read-only storage object Read data from the object.

According to the first data access method for a storage system according to the eighteenth aspect of the present invention, there is provided a second data access method for a storage system according to the eighteenth aspect of the present invention, wherein the storage object includes data from the first storage device. The first storage space and the second storage space from the second storage device.

According to the first data access method for a storage system according to the eighteenth aspect of the present invention, there is provided a third data access method for a storage system according to the eighteenth aspect of the present invention, wherein the storage object includes data from the first storage device. The portion of the first contiguous storage space and the portion of the second contiguous storage space from the second storage device.

According to the second or third data access method for a storage system according to the eighteenth aspect of the present invention, there is provided a fourth data access method for a storage system according to the eighteenth aspect of the present invention, wherein the second storage space is used for Verification data of the data in the first storage space is stored.

According to the first data access method for a storage system according to the eighteenth aspect of the present invention, there is provided a fifth data access method for a storage system according to the eighteenth aspect of the present invention, wherein the storage object includes storage space from a storage device .

According to the first data access method for a storage system according to the eighteenth aspect of the present invention, there is provided a sixth data access method for a storage system according to the eighteenth aspect of the present invention, further comprising recording the written data and the A mapping of writable objects.

A sixth data access method for a storage system according to the eighteenth aspect of the present invention provides a seventh data access method for a storage system according to the eighteenth aspect of the present invention, further comprising responding to a read request, according to the The mapping relationship searches for a read-only storage object that stores the requested data, and reads data from the read-only storage object that stores the requested data.

According to the aforementioned data access method for a storage system according to the eighteenth aspect of the present invention, there is provided an eighth data access method for a storage system according to the eighteenth aspect of the present invention, wherein if the writable object is full, further The free storage object is set as a writable storage object, so that the storage system includes at least one writable storage object.

According to the aforementioned data access method for a storage system according to the eighteenth aspect of the present invention, there is provided a ninth data access method for a storage system according to the eighteenth aspect of the present invention, wherein the storage system further comprises an NVDIMM; the The method further includes: in response to a write request, writing data to the NVDIMM; in response to completion of the operation of writing data to the NVDIMM, sending a message indicating completion of the write request; and in response to writing data to the available Write the storage object to release the space occupied by the data in the NVDIMM.

According to the aforementioned data access method for a storage system according to the eighteenth aspect of the present invention, there is provided a tenth data access method for a storage system according to the eighteenth aspect of the present invention, wherein the storage system further comprises an NVDIMM; the The method further includes: in response to the write request, writing data to the NVDIMM; in response to the completion of the operation of writing data to the NVDIMM, sending a message indicating that the write request is completed; after setting the writable storage object to The data is read from the NVDIMM in response to a read request before the read-only storage object.

According to the first to ninth data access methods for a storage system according to the eighteenth aspect of the present invention, there is provided an eleventh data access method for a storage system according to the eighteenth aspect of the present invention, wherein the storage system further A cache is provided, and the method further includes: in response to a write request, writing data to the cache; before setting the writable storage object as a read-only storage object, in response to a read request, reading out from the cache the data.

According to an eighteenth aspect of the present invention, there is provided a twelfth data access method for a storage system according to the eighteenth aspect of the present invention, wherein the storage system includes a plurality of storage devices and NVDIMMs, and the storage system provides multiple storage objects, the storage objects are composed of storage resources on the storage device, the multiple storage objects include one or more writable storage objects and multiple read-only storage objects; the method includes: in response to the first write request, writing first data to the NVDIMM; in response to writing the first data to the NVDIMM, sending a message indicating that the first write request is complete; in response to the second write request, writing the second data to all the NVDIMM; in response to writing the second data to the NVDIMM, sending a message indicating that the second write request is complete; generating a storage data block, the storage data block including the first data and the second data; write the storage data block to the writable storage object by appending/sequential writing (append); if the writable storage object is full, set the writable storage object as a read-only storage object ; in response to a read request, read the first data or the second data from the read-only storage object.

According to a nineteenth aspect of the present invention, there is provided a first computer according to the nineteenth aspect of the present invention, comprising: a machine-readable memory for storing program instructions; one for executing the program instructions stored in the memory or more processors; the program instructions are used to cause the one or more processors to execute the various methods provided according to the eighteenth aspect of the present invention.

According to a twentieth aspect of the present invention, there is provided a first data access apparatus for a storage system according to the twentieth aspect of the present invention, the storage system comprising a plurality of storage devices, the storage system providing a plurality of storage objects, The storage object is composed of storage resources on the storage device, and the multiple storage objects include one or more writable storage objects and multiple read-only storage objects; the device includes: a writing module for responding to a writing request, Write data to the writable storage object in an append/sequential writing manner; a storage object setting module is used to set the writable storage object as read-only storage if the writable storage object is full an object; a readout module, configured to read data from the read-only storage object in response to a read request.

According to a twentieth aspect of the present invention, there is provided a second data access apparatus for a storage system according to the twentieth aspect of the present invention, wherein the storage system includes a plurality of storage devices and NVDIMMs, and the storage system provides a plurality of a storage object, the storage object is composed of storage resources on a storage device, the multiple storage objects include one or more writable storage objects and multiple read-only storage objects; the apparatus includes: a first NVDIMM writing module for In response to the first write request, write first data to the NVDIMM; a first message sending module is configured to send a message indicating that the first write request is completed in response to the first data being written to the NVDIMM a second NVDIMM writing module for writing second data into the NVDIMM in response to a second writing request; a second message sending module for sending an instruction to the NVDIMM in response to writing the second data into the NVDIMM a message that the second write request is completed; a generating module is used to generate a storage data block, the storage data block includes the first data and the second data; a storage object writing module is used to write to the writable The storage object is written into the storage data block in an append write/sequential write (append) manner; a storage object setting module is used to set the writable storage object as a read-only storage object if the writable storage object is full ; a readout module, configured to read out the first data or the second data from the read-only storage object in response to a read request.

Description of drawings

The present invention, together with the preferred mode of use and its further objects and advantages, will be best understood by reference to the following detailed description of illustrative embodiments when read in conjunction with the accompanying drawings, which include:

1 is a block diagram of a prior art storage system;

2 is a block diagram of a storage system according to an embodiment of the present invention;

3 is a block diagram of a storage system according to yet another embodiment of the present invention;

4 is a flowchart of a data writing method according to an embodiment of the present invention;

FIG. 5 illustrates a data block stored on an NVDIMM according to an embodiment of the present invention;

FIG. 6 is a flowchart of a method executed during a boot process of a storage system according to an embodiment of the present invention;

7 is a flowchart of a data writing method according to another embodiment of the present invention;

FIG. 8 illustrates the data organization of an NVDIMM according to an embodiment of the present invention;

9 is a flowchart of a storage space allocation method for an NVDIMM according to an embodiment of the present invention;

10 is a flowchart of a storage space allocation method for an NVDIMM according to yet another embodiment of the present invention;

11 is a flowchart of a method for releasing storage space of an NVDIMM according to an embodiment of the present invention;

12 is a flowchart of a method for releasing storage space of an NVDIMM according to another embodiment of the present invention;

13 is a flowchart of a data writing method according to yet another embodiment of the present invention;

Figure 14 illustrates a storage object according to an embodiment of the present invention;

15 shows a schematic diagram of read and write operations of a storage system according to an embodiment of the present invention;

16 is a flowchart of a storage system data access method according to an embodiment of the present invention;

FIG. 17 is a flowchart of a data access method for a storage system according to yet another embodiment of the present invention; and

FIG. 18 is a flowchart of a data access method of a storage system according to still another embodiment of the present invention.

Detailed ways

2 is a block diagram of a storage system according to an embodiment of the present invention. In the embodiment according to FIG. 2 , the storage system may be a computer or server, including a CPU 210 , NVDIMM 220 , and one or more disk (DISK) devices 230 . The disk device 230 may be a mechanical hard disk, a solid state disk, and/or a memory card. The disk device 230 can exchange data with the CPU through, for example, SATA, IDE, USB, PCIe, NVMe, SCSI, Ethernet, and the like. The disk device 230 may be directly coupled to the CPU 210, or may be coupled to the CPU 210 through a bridge chip (not shown) such as a chipset. NVDIMM 220 is coupled to CPU 210 through a DIMM memory socket. CPU 210 may include one or more CPU chips. One or more disks 230 may be organized in RAID to provide high performance, high reliability storage services.

Software that accesses the storage system, such as application software, database software, and the like, runs on the CPU 210 . Software for providing storage services according to embodiments of the present invention also runs on the CPU 210 for responding to storage requests and operating storage devices such as the NVDIMM 220 and the disk 230 . The invention can thus be implemented on servers such as application servers, database servers.

3 is a block diagram of a storage system according to yet another embodiment of the present invention. In the embodiment according to Figure 3, a separate storage system is provided. The storage system includes a controller 310 , an NVDIMM 320 , an interface 340 , and one or more disk (DISK) devices 330 . The disk device 230 may be a mechanical hard disk, a solid state disk, and/or a memory card. The disk device 230 may exchange data with the controller through, for example, SATA, IDE, USB, PCIe, NVMe, SCSI, Ethernet, and the like. The disk device 330 may be directly coupled to the controller 310, or may be coupled to the controller 310 through a bridge chip (not shown) such as a chipset. NVDIMM 320 is coupled to controller 310 through a DIMM memory socket. Controller 10 may include one or more CPU chips, or one or more application specific integrated circuits. One or more disks 330 may be organized in RAID to provide high performance, high reliability storage services. Interface 340 couples the storage system to the network through which the storage system can be accessed. The interface 340 may be an interface supporting, for example, Ethernet, FC (Fibre Channel, Fibre Channel), and Infiniband. Applications accessing the storage system run on other servers and access the storage system according to the embodiment of FIG. 3 through the network and the interface 340 . The software for providing storage services according to the embodiment of the present invention runs on the controller 310 to respond to storage requests and operate storage devices such as the NVDIMM 320 and the disk 330 .

FIG. 4 is a flowchart of a data writing method according to an embodiment of the present invention. When data needs to be written, the application issues a data write request. The software according to the embodiment of the present invention receives a data write request sent by an application program or other program. In response to receiving the data write request (410), in an embodiment according to the invention, data is written to the NVDIMM (420). After the data is written to the NVDIMM, a message is issued to the application or other program that issued the write request to indicate that the data write request has been processed (430). Although data is only written into the NVDIMM at this time, and has not been written to the storage device of the storage system, such as the disk 230 (see FIG. 2 ), since the NVDIMM has non-volatile characteristics, the writing method according to the embodiment of the present invention In the NVDIMM, a message indicating that the data write request has been processed is issued based on the data being written to the NVDIMM, and subsequent steps ensure that the data written to the NVDIMM is written to a storage device such as disk 230 of the storage system. In this way, it can be ensured that even if a power failure occurs between executing step 430 and executing step 450, the data written to the storage system is not lost. In the embodiment according to Figure 4, the NVDIMM acts as a write cache for the storage server. Since the NVDIMM has high-speed data access capability, step 420 will be completed quickly, and a message indicating the completion of the data write request can be sent to the NVDIMM, thereby improving the write performance of the storage system.

In response to receiving the data write request, in an embodiment according to the invention, the data is also written to the storage device (450). In an embodiment according to the invention, both the steps of writing data to the storage device (450) and the steps of writing data to the DIMM (420) occur in dependence on receiving a data write request (410). Thus, in one example, the step of writing data to the storage device (450) occurs in parallel with the step of writing data to the DIMM (420). In another example, the step of writing data to the storage device (450) is handled by one CPU and the step of writing data to the DIMM (420) is handled by another CPU. In yet another example, the steps of writing data to the storage device (450) and the steps of writing data to the DIMMs (420) are time-shared by one CPU. In yet another example, the step of writing data to the DIMM (420) is performed after the step of writing the data to the storage device (450).

In response to the completion of the step of writing data to the storage device (450), it is instructed to release the space occupied in the NVDIMM by the data of the NVDIMM written in step 420. The storage device may be a storage device such as disk 230 (see FIG. 2). Since the data cached in the NVDIMM is released or deleted in time after the data is written to the storage device, in the embodiment according to the present invention, the storage capacity of the NVDIMM does not need to be large, and can be much smaller than that of the storage system such as the disk 230 (see FIG. 2) The storage capacity of the storage device. The release of data in the NVDIMM may be to mark the storage space occupied by the data as free without performing delete, write or erase operations on the NVDIMM. In the embodiment according to Figure 4, in response to receiving a data write request, data is written to the storage device (450) and corresponding data is released from the NVDIMM (460). Therefore, data will not be cached in the NVDIMM for a long time, so that the storage space of the NVDIMM can be quickly reused, reducing the overall storage capacity requirement of the NVDIMM. And make the storage system have sustainable high write performance.

Those skilled in the art will appreciate that multiple data write requests may be received. For each data write request, the step of writing data to the NVDIMM (420) and the step of writing data to the storage device (450) are performed separately. And after writing the data to the NVDIMM (420), send a message (430) indicating that the data write request is complete. After the data is written to the storage device (450), the space occupied by the data in the NVDIMM is released (460).

Figure 5 illustrates a block of data stored on an NVDIMM according to an embodiment of the present invention. In the embodiment according to FIG. 5 , a data block 510 , a data block 520 , a data block 530 and a data block 540 are stored on the NVDIMM 500 . As an example, the data block 510 includes at least data, a logical address and a sequence number. In one example, data block 510, data block 520, data block 530 and data block 540 have the same size. The data block 510, the data block 520, the data block 530 and the data block 540 may also have different sizes, and the respective sizes of the data blocks are recorded in each data block.

Referring to FIG. 4 , the received data write request includes the data to be written and the logical address of the data to be written. In order to write data to the NVDIMM, a data block 510 ( FIG. 5 ) is generated, and the data to be written and the logical address to be written as the data included in the data write request are recorded in the data block 510 . When the data block 510 is generated, a sequence number is also generated and included in the data block 510 . The sequence number is incremented and is used to identify the sequence in which the data write request is received, so that the sequence in which each data block is generated can be obtained through the sequence number. To obtain the incremented sequence number, in one example, whenever a data write request is received and a data block (eg, data block 510 ) is generated, the sequence number is incremented, and the value of the incremented sequence number is recorded in the data in block 510.

When data needs to be recovered from the NVDIMM, a data write request to write data to the logical address can be obtained by using the data and the logical address in the data block (eg, data block 510 ); and by using the sequence number in the data block, it can be Gets the order in which each data write request was received. It is important to accurately identify the order in which individual data write requests are received. Data write requests that occur at different times may write different data to the same address, and the execution order of the data write requests determines the data finally recorded at the same address.

Various metadata are also recorded in the NVDIMM. Metadata can be used to record free and/or occupied memory areas in the NVDIMM. When writing a data block to the NVDIMM, the data block is written to a free storage area in the NVDIMM, and the storage area to which the data is written is marked as occupied. When the space occupied by the data in the NVDIMM (see Figure 4, 460) is freed in response to writing data to a storage device such as disk 230 (see Figure 2), the corresponding storage area is marked as free in the metadata. When writing data to the NVDIMM, if there is not enough free memory area in the NVDIMM, the data writing cannot be completed. The data writing can be temporarily suspended, and the operation of writing data to the NVDIMM can be performed after the occupied storage area in the NVDIMM is released.

Metadata can also be used to record flag bits that indicate whether data recovery operations are performed when the storage system is powered up. In one example, the flag bit is set to a first value when the storage system starts up, and the flag is set to a second value when the storage system shuts down normally. In this case, when the storage system starts up, if the flag bit is found to be the first value, it indicates that the storage system is not properly shut down. In this case, data recovery needs to be performed using the data saved in the NVDIMM. The method for recovering data from the NVDIMM will be described in detail later with reference to the accompanying drawings. In another example, when the storage system is shut down abnormally, the first value is written in the flag bit. When the storage system starts up, if the flag bit is found to be the first value, it indicates that the storage system has not been shut down properly, and the NVDIMM needs to be used Save the data for data recovery.

FIG. 6 is a flowchart of a method performed during a boot process of a storage system according to an embodiment of the present invention. When the storage system is powered on (600), a flag bit recorded in the NVDIMM is accessed (620) to determine whether the storage system needs to restore data from the NVDIMM. At a graceful shutdown, the data cached in the NVDIMM has all been written to a storage device such as disk 230 as an example. In this case, there is no need to restore data from the NVDIMM. exist. If the storage system experiences an abnormal shutdown, such as an unexpected event such as a power failure during the execution of a data write request, the data corresponding to the data write request is written to the NVDIMM, but not written to the storage device such as the disk 230 middle. In this case, data needs to be recovered from the NVDIMM.

It is determined by accessing the flag bit recorded in the NVDIMM whether the storage system experienced a normal shutdown during the last shutdown (630). In the case of normal shutdown, there is no need to restore data from the NVDIMM, and the method in FIG. 6 turns to step 660 to end the execution. In the case of an abnormal shutdown, all data blocks are read from the NVDIMM (640). In one example, referring to FIG. 5, data block 510, data block 520, data block 530, and data block 540 are read from the NVDIMM. In a further embodiment, the information of the occupied data blocks is also obtained by accessing the metadata recorded in the NVDIMM, and then the occupied data blocks are read out from the NVDIMM without reading out the free data blocks in the NVDIMM.

The read data blocks are sorted in ascending order according to the sequence numbers recorded in each data block, and the data corresponding to each data block is written to a storage device such as the disk 230 according to the sorted order. The sequence number of the data block is small, which means that the data write request corresponding to the data block occurs earlier. If the sequence number of the data block 510 (see FIG. 5 ) is smaller than that of the data block 520 , it means that the data write request corresponding to the data block 510 occurs earlier than the data write request corresponding to the data block 520 . In one example, the sequence number of the data block 510 is smaller than the sequence number of the data block 520, the sequence number of the data block 520 is smaller than the sequence number of the data block 530, and the sequence number of the data block 530 is smaller than the sequence number of the data block 540, according to the data In the sequence of block 510, data block 520, data block 530, and data block 540, the data corresponding to each data block is written to a storage device such as disk 230 (see FIG. 2). Specifically, data and logical addresses are obtained from each data block, and data is written to a storage device such as disk 230 according to the logical addresses. In one example, the data write request is regenerated based on the logical address and the data, and the data is written to the storage device based on the data write request.

FIG. 7 is a flowchart of a data writing method according to yet another embodiment of the present invention. When data needs to be written, the application issues a data write request. The software according to the embodiment of the present invention receives a data write request sent by an application program or other program. In another embodiment, the storage system according to the present invention receives data write requests sent by application programs or other programs or other servers through a network.

In response to receiving the first data write request (710), in the embodiment according to FIG. 7, the first data is written to the NVDIMM (712). The first data write request includes first data and a first logical address to which the first data is to be written. After the first data is written to the NVDIMM, a message is sent to the application, other program, or server that issued the write request, the message indicating that the first data write request has been processed (714). Although data is only written to the NVDIMM at this time, and has not been written to the storage device of the storage system, such as the disk 230 (see FIG. 2 ), since the NVDIMM has non-volatile characteristics, the writing method according to the embodiment of the present invention , a message indicating that the data write request has been processed is issued based on the data being written to the NVDIMM, and subsequent steps ensure that the data written to the NVDIMM is written to a storage device such as disk 230 of the storage system.

In the embodiment according to FIG. 7 , after the first data write request is received, the first data corresponding to the first write request is not written to the storage device such as the disk 230 immediately. Instead, wait for a second write request to be received (720). By combining the first write request and the second write request, the number of write operations performed by the storage device can be reduced on the premise of ensuring data reliability, thereby improving the performance of the storage system. The second data write request includes second data and a second logical address to which the second data is to be written.

In response to receiving the second data write request (720), second data is written to the NVDIMM (722). After writing the second data to the NVDIMM, a message is sent to the application, other program, or server that issued the write request to indicate that the second data write request has been processed (724).

In response to receiving the first write request and the second write request, the first data and the second data are written to the storage device (730). In one example, a storage data block is generated, the first data and the second data are recorded in the storage data block, and the storage data block is written into a storage device. The storage data block may have the same size as the physical storage block of the storage device. In another example, when the first data write request and the second data write request are consecutive in logical addresses, the second data is appended to the first data and written to the storage device.

In response to writing the first and second data to the storage device, the first and second data are released in the NVDIMM (740). In one example, a first memory block corresponding to the first data and a second memory block corresponding to the second data are released in the NVDIMM.

It should be noted that the data merged in step 730 may come from two or more data write requests. In another example, after receiving the second data write request, if the first data and the second data are not suitable for combining, the first data and the second data are written into the storage device respectively. In the embodiment according to FIG. 7 , the memory blocks written to the NVDIMM may be memory blocks as shown in FIG. 5 . After the first data and the second data are written into the NVDIMM according to the embodiment of FIG. 7 , if an abnormal power failure is experienced, the data can also be recovered from the NVDIMM by the embodiment shown in FIG. 6 .

Figure 8 illustrates the data organization of an NVDIMM according to an embodiment of the present invention. Referring to Figure 8, the storage space in the NVDIMM is organized in a manner similar to the ring buffer (810). The head of the queue is indicated in Figure 8, indicating the starting position where data is normally written to the NVDIMM. The queue tail is also indicated in FIG. 8 . The unused or released storage space in the NVDIMM is free storage space. As shown in FIG. 8 , the NVDIMM includes free storage space 812 , free storage space 814 and free storage space 816 . Use a data structure to index free storage space in an NVDIMM. In the embodiment of FIG. 8, the free storage space of the NVDIMM is organized in a linked list 830, for example. Data structures such as linear tables, trees, etc. can also be used to index free storage space in NVDIMMs. The NVDIMM also includes one or more data blocks, which are identified by mesh lines in Figure 8. A data block represents the occupied storage space in an NVDIMM.

Referring to FIG. 8 , the linked list 830 includes a node 832 , a node 834 and a node 836 . Each node indexes contiguous free address space in the NVDIMM. Node 832 indexes free storage space 812 , node 834 indexes free storage space 814 , and node 836 indexes space storage space 816 . In node 836, free memory space 816 is indexed by storing the start address, length, and/or end address of free memory space 816. In node 834, free memory space 814 is indexed by storing the start address, length, and/or end address of free memory space 814. The space storage space 812 is indexed in the node 832 by storing the start address, length and/or end address of the free storage space 812 . Node 832, node 834 and node 834 are organized as a doubly linked list. Pointer 820 is also provided to index node 836. Referring to Figure 8, sort the free storage space of the NVDIMM in a clockwise direction. In the absence of address wrapping (wrap refers to addresses reaching or past the maximum value and starting over from the beginning of the address space), nodes 832, 834 and 834 are ordered by the addresses of the free memory regions they each index.

Free storage space 816 is adjacent to free storage space 814 , which is adjacent to free storage space 812 . Referring to FIG. 8 , there are data blocks between the free storage space 816 and the free storage space 812 or the free storage space 814 , so the free storage space 816 is not connected to the free storage space 812 or the free storage space 814 .

When free storage space is allocated or allocated from the NVDIMM, the free storage area is searched from the node 836 indexed by the pointer 820 . According to the storage space allocation/release method according to the embodiment of the present invention, a node indexed by the pointer 836 has a high probability of having a free storage area that meets the requirements, thereby improving the efficiency of storage space allocation. In one example, when a free storage area 816 that meets the requirements is found at node 836, the first address of the free storage area 816 is returned to represent the allocated storage space. Modify the start address and/or length of the free memory area recorded in node 836. At this time, the head of the queue changes accordingly. As another example, when a free storage area that meets the requirements cannot be found at the node 836, the linked list 830 is traversed to find a free storage area that meets the requirements. For example, if a free storage space that meets the requirements is found at the node 832, the storage space is allocated from the free storage area 812 indexed by the node 832, and the index information of the node 832 is modified accordingly.

Typically, data is written from the head (head) of the NVDIMM and data is deleted from the tail (tail). Referring to Figure 4, after data is written to the storage device (450), the space occupied by the data in the NVDIMM is released (460). Therefore, the data in the NVDIMM is released after being written to the NVDIMM for a period of time, and when data is written to the NVDIMM, a free storage area that meets the requirements will be found at the node 836 indexed by the pointer 820 under normal circumstances.

In another example according to the present invention, a certain space is reserved in the NVDIMM to store state information of the NVDIMM, including, for example, various metadata. Metadata can be used to record free and/or occupied memory areas in the NVDIMM. Metadata can also be used to record flag bits that indicate whether data recovery operations are performed when the storage system is powered up. Those skilled in the art will appreciate that metadata may also be stored outside of the NVDIMM.

FIG. 9 is a flowchart of a storage space allocation method of an NVDIMM according to an embodiment of the present invention. In the embodiment according to FIG. 9, when writing data to the NVDIMM, free memory space needs to be allocated from the NVDIMM. Referring to FIG. 8, one or more free storage spaces (812, 814, and 816) may be included in the NVDIMM, and a linked list 830 or other data structure is used to index the one or more free storage spaces in the NVDIMM.

In response to receiving a request to allocate storage space (910), free storage space is looked up (920) from node 836 (the first node) indexed by pointer 820 (see Figure 8). The free storage space 816 indexed by the node 836 is used as the queue head of the ring buffer of the NVDIMM, and there is a high probability that there is a free buffer that can meet the allocation requirements. If the first node can satisfy the request to allocate storage space (930), free storage space is allocated from the first node (940). In one example, the first node's free storage starting address is provided in response to the request to allocate storage. If the first node cannot satisfy the request for allocating storage space (930), it traverses other nodes of the data structure such as linked list 830 to find free storage space (950). For example, if the free storage space 816 corresponding to the node 836 in FIG. 8 cannot meet the request for allocating storage space, the linked list 830 is traversed to determine whether the free storage space 814 corresponding to the node 834 meets the request for allocating storage space. If necessary, it is determined whether the free storage space 812 corresponding to the node 832 satisfies the request for allocating storage space. In a further embodiment, if the free storage space that satisfies the request cannot be found after traversing the linked list 830, then wait for the free storage space that appears in the NVDIMM due to the release of the storage space. In still further embodiments, the release of storage space occurs most likely in the vicinity of free storage space 816, and thus, upon finding that the storage space in the NVDIMM is freed, node 836 is prioritized to search for free storage space.

In the embodiment according to the present invention, if the free storage space 816 that satisfies the allocation request is found through the node 836, the node 836 is also modified to index the free storage space after the storage space is allocated. For example, the start address, end address, and/or length of the free memory space stored in node 836 is modified.

FIG. 10 is a flowchart of a storage space allocation method for an NVDIMM according to yet another embodiment of the present invention. In the embodiment according to FIG. 9 , the linked list 830 is traversed to find the storage space that satisfies the allocation request from all the free storage space of the NVDIMM. According to the present invention, there is a higher probability of free storage space at the head of the queue of the ring buffer of the NVDIMM compared to other positions in the buffer. Therefore, in the embodiment according to FIG. 10 , the free storage space is only searched from the first node (node 836 ) in the queue head of the index ring buffer, and when the free storage space corresponding to the first node cannot satisfy the allocation request, no free storage space is required. To traverse the linked list 830, instead wait for free storage space to appear that satisfies the allocation request. Thereby, the efficiency of storage space allocation is further improved.

Referring to Figure 10, in response to receiving a request to allocate storage space (1010), free storage space is looked up (1020) from node 836 (the first node) indexed by pointer 820 (see Figure 8). The free storage space 816 indexed by the node 836 is used as the queue head of the ring buffer of the NVDIMM, and there is a high probability that there is a free buffer that can meet the allocation requirements. If the first node can satisfy the request to allocate storage space (1030), free storage space is allocated from the first node (1040). In one example, the first node's free storage starting address is provided in response to the request to allocate storage. If the first node cannot satisfy the request for allocating storage space (1030), it waits for a free storage space that satisfies the allocation request to appear (1050). In one example, in response to the release of storage space for the NVDIMM, it is again looked up whether there is free storage space in node 836 (the first node) that satisfies the allocation request (1040). In another example, in response to the storage space release of the NVDIMM, it is determined whether the storage space released by the storage space release request can satisfy the allocation request, and then the free storage space is allocated.

FIG. 11 is a flowchart of a method for releasing storage space of an NVDIMM according to an embodiment of the present invention. In response to receiving the request to release the first storage space (1110), it is determined whether the first storage space and the adjacent free storage space are connected in the NVDIMM (1120). If the first storage space is connected to the adjacent storage space in the NVDIMM, the first storage space and the adjacent free storage space are merged (1130). If the first storage space is not contiguous with adjacent free storage spaces in the NVDIMM, a new node is provided to index the first storage space (1140). For example, referring to FIG. 8, a new node is created to index the freed first storage space, and the new node is inserted into the linked list 830. The new node may be inserted into the linked list 830 according to the address of the first storage space. The nodes in the linked list 830 are still sorted according to the addresses of the free storage spaces indexed by them. In other words, the free storage spaces indexed by each node in the linked list 830 are arranged in a clockwise direction according to their addresses.

In an embodiment according to the present invention, the nodes of the linked list 830 are ordered by the addresses of the free storage spaces they index. According to the address of the first storage space, the node indexing the free storage space adjacent to the first storage space is found in the linked list 830, and it is determined whether the first storage space and the adjacent free storage space are connected. In one example, the start address, end address and/or length of the free memory space are included in the node indexing the free memory space. Whether the first storage space and the adjacent free storage space are connected is determined by comparing the address of the free storage space with the address of the first storage storage space.

In one example, referring to FIG. 8, if the first storage space to be released is adjacent to the free storage space 816, and the end address of the first storage space is equal to the start address of the free storage space 816, then the modified index of the free storage space Node 836 of 816 , so that the modified node 836 indexes the first storage space and the free storage space 816 . In this way, the merging of the first storage space and the free storage space 816 is achieved. For example, the start address, length and/or end address of the free storage space indexed by the node 836 is modified, so that the node 836 indexes the combined first storage space and the free storage space 816 . In another example, the first storage space and the free storage space 816 are merged by creating a new node that indexes the first storage space and the free storage space 816 , inserting the new node into the linked list 830 , and deleting the node 836 . In this example, since the node 836 pointed to by the pointer 820 was deleted, the pointer 820 also points to the new node that was created.

In still another example, if the free storage space adjacent to the first storage space to be released is the free storage space 816 and the free storage space 812, but the first storage space is not the same as the free storage space 816 or the space storage space 812 Then, a new node is created to index the first storage space, and the new node is inserted into the linked list 830 .

Those skilled in the art will appreciate that, although in the embodiment of FIG. 8, a linked list is used to organize multiple nodes indexing free storage space, multiple data structures can be used to organize multiple nodes, such as trees, linear lists, etc. and other structures to organize multiple nodes and sort and/or efficiently search multiple nodes.

FIG. 12 is a flowchart of a method for releasing storage space of an NVDIMM according to yet another embodiment of the present invention. In response to receiving the request to release the first storage space (1210), look up whether the forward contiguous free storage space of the first storage space exists (1220), in other words, look for the free space before the first storage space in the NVDIMM Whether the storage space is connected to the first storage space. If the free storage space connected to the front is found, then look for the free storage space connected to the back of the first storage space (1230), in other words, find the free storage space after the first storage space in the NVDIMM and the first storage space. Whether the storage space is connected. If the back-connected free storage space is found in operation 1230, indicating that there are free storage spaces connected to the first storage space before and after the first storage space in the NVDIMM, then the free storage space connected to the first storage space in the forward direction will be The storage space and the free storage space adjoining the first storage space backwards are merged (1250). In one example, the merged free storage space is indexed by a node in the linked list 830 .

If no free storage space connected to the back of the first storage space is found in operation 1230, it indicates that there is a free storage space connected to the first storage space before the first storage space in the NVDIMM, and there is no free storage space after the first storage space. If there is free storage space adjoining the first storage space, then the free storage space adjoining the first storage space forward and the first storage space are merged (1260). In one example, the node in the linked list 830 that indexes the free storage space that is forward-connected to the first storage space is modified to index the combined free storage space.

In operation 1220, if the forward contiguous free storage space is not found, then look up whether the backward contiguous free storage space of the first storage space exists (1235), in other words, look up the space after the first storage space in the NVDIMM. Whether the free storage space is connected to the first storage space. If a backward-adjacent free storage space is found, the first storage space and the backward-adjacent free storage space are merged (1270). In one example, the node in the linked list 830 that indexes the free storage space that is backward adjacent to the first storage space is modified to index the combined free storage space.

If no free storage space adjacent to the first storage space is found in operation 1235, it indicates that there is no free storage space around the first storage space in the NVDIMM. The new node is then provided to index the first storage space (1280), and the new node is inserted into the linked list 830 according to the address of its indexed first storage space, so that the nodes in the linked list 830 are sorted by the addresses of their respective indexed free storage areas.

FIG. 13 is a flowchart of a data writing method according to yet another embodiment of the present invention. When data needs to be written, the application issues a data write request. The software according to the embodiment of the present invention receives a data write request sent by an application program or other program. In another embodiment, the storage system according to the present invention receives data write requests sent by application programs or other programs or other servers through a network. NVDIMMs are used as write caches in the data writing method according to the present invention.

In response to receiving the data write request (1310), in the embodiment according to FIG. 13, it is determined whether there is enough free storage space in the NVDIMM to accommodate the data corresponding to the data write request (1320). If there is enough free storage space in the NVDIMM, allocate the storage free storage space in the NVDIMM (1330), write the data corresponding to the data write request into the NVDIMM (1350), and then write the data to the application that issued the data write request. , other program or server issues a message to indicate that the data write request has been processed (1360). Although data is only written into the NVDIMM at this time, and has not been written to the storage device of the storage system, such as the disk 230 (see FIG. 2 ), since the NVDIMM has non-volatile characteristics, the writing method according to the embodiment of the present invention , a message indicating that the data write request has been processed is issued based on the data being written to the NVDIMM, and subsequent steps ensure that the data written to the NVDIMM is written to a storage system such as disk 230 or disk 330 (see FIG. 3) storage device.

If there is not enough contiguous free memory space in the NVDIMM to satisfy the data write request, then wait for free memory space to appear in the NVDIMM (1340). As storage space is released to the NVDIMM, a larger contiguous free storage space will appear in the NVDIMM to satisfy data write requests.

In response to receiving the data write request (1310), data corresponding to the data write request is written to the storage device (1370). In response to the completion of the step of writing the data to the storage device (1370), it is instructed to release the space occupied by the data corresponding to the data writing request in the NVDIMM (1380). The storage device may be a storage device such as disk 230 (see FIG. 2). As the space occupied by data in the NVDIMM is freed, there will be enough contiguous free storage space in the NVDIMM to satisfy the need to allocate storage space in the NVDIMM. In another example, storage space may be freed from the NVDIMM by the embodiments shown in FIG. 11 or FIG. 12 .

In the embodiment according to Figure 13, in response to receiving a data write request, data is written to the storage device (1370) and corresponding data is released from the NVDIMM (1380). Therefore, the data will not be cached in the NVDIMM for a long time, so that the storage space of the NVDIMM can be quickly reused, reducing the demand for the overall storage capacity of the NVDIMM, and applying for free storage space in the NVDIMM in time to respond to data. write request.

Those skilled in the art will appreciate that multiple data write requests may be received. For each data write request, the steps of writing data to the NVDIMM (1350) and the steps of writing data to the storage device (1370) are performed, respectively. And after writing the data to the NVDIMM (1350), send a message (1360) indicating that the data write request is complete. After the data is written to the storage device (1370), the space occupied by the data in the NVDIMM is released (1380).

Referring to FIG. 8 again, in response to receiving the data write request, the node 836 is obtained through the pointer 820, and at least part of the free memory area is allocated from the free memory area 816 corresponding to the node 836 for writing data. After data is written, free memory area 816 will become small and may not be sufficient to respond to new data write requests. According to the data writing method of the embodiment of the present invention, the data block at the tail of the queue of the NVDIMM (indicated by tail) is written at an earlier time, so in general, the data block will be written soon with a high probability into the storage device and released in the NVDIMM. That is, the data block at the tail of the queue (indicated by tail) will be freed soon. As the data blocks at the tail of the queue (indicated by tail) are freed, the free storage area (816) indexed by node 836 will be merged with the data blocks at the tail of the queue (indicated by tail), making the free storage area 816 larger , and be able to respond to new data write requests.

In the embodiment according to the present invention, when data is written into the storage device, the written data is formed into a data stream, and the distribution of the written data on the storage device is localized, and the data to be read is distributed in the The entire storage space of the storage device. Through this data part method, the advantages of SSD random read performance and sequential write performance are fully utilized.

Figure 14 illustrates a storage object according to an embodiment of the present invention. In an embodiment according to the present invention, the storage resources provided by the storage device are organized into one or more storage objects. In one example, storage resources (eg, blocks) of an SSD are pooled and organized into one or more storage objects. Storage objects are also called Containers. When data is written to or read from a storage device, the storage object is used as the basic unit of access. The number of storage objects is determined by the capacity of the storage system.

Referring to FIG. 14, Disk 0, Disk 1, Disk 2, and Disk 3 are storage devices such as SSDs. In other examples, Disk 0, Disk 1, Disk 2 and Disk 3 may also be mechanical hard disks. As an example, Disk 0 provides the address space as a storage device. The address space of disk 0 is divided into a plurality of chunks, such as chunk 0 1410, chunk 1 1412 . . . chunk n in FIG. 4 . A large block may include multiple blocks. Similarly, the address space of each of Disk 1, Disk 2, and Disk 3 is divided into a plurality of large blocks, eg, large blocks 1420 and 1422, are provided by Disk 3.

In the example of FIG. 14, a plurality of chunks provided by Disk 0, Disk 1, Disk 2, and Disk 3 form a storage resource pool. Several large blocks are organized into storage objects (shown as containers in Figure 14) from the storage resource pool. For example, container 0 includes chunk 0 (chunk 0-1) from disc 0, chunk 1 (chunk 1-1) from disc 1, and chunk 0 (chunk 2-0) from disc 2, Container 1 includes chunk 2 from disc 0 (chunk 0-2) and chunk 2 from pan 1 (chunk 1-2). A storage system according to an embodiment of the present invention provides a plurality of containers, each container may include the same or a different number of chunks. Within the container, data protection mechanisms such as RAID can be provided. For example, in container 0, chunks 0-0 and 1-1 are used to store user data, while chunks 2-0 are used to store checksums corresponding to chunks 0-0 and 1-1 data. In container 0, the storage location of the verification data is not limited to the large block 2-0, and can also be stored in the large block 0-0 or the large block 1-1. As another example, both chunks 0-2 and 1-2 of container 1 are used to store user data. Still as an example, the multiple chunks that make up container 0 (chunk 0-0, chunk 1-1, and chunk 2-0) are from different disks, so that access to container 0 will be distributed across disk 0, 1 and disc 2. In this way, the parallelism of access to the disk is increased, resulting in improved storage performance. Those skilled in the art will also appreciate that the disks in a storage system may be of different capacities, and the disks may be replaced so that large blocks within the container may be from the same disk.

In the example of FIG. 14, the example in which a container is comprised by a large block is shown. Chunks may be constructed in other ways. For example, in another example, instead of providing large blocks, the container corresponds to a segment of storage space on the disk, eg, a segment of contiguous logical address space.

In the embodiment according to the present invention, in order to improve the performance of the storage system, when data is written to the container, an additional write or sequential write is used. In other words, the container can only be written from the beginning to the end, and any part of the container cannot be overwritten.

FIG. 15 shows a schematic diagram of read and write operations of a storage system according to an embodiment of the present invention. The storage system shown in FIG. 15 includes container 0, container 1, container 2 and container 3. Containers in storage systems can be divided into two categories, writable containers and read-only containers. In Figure 15, container 3 is a writable container, while container 0, container 1 and container 2 are all read-only containers. In the storage system, there is at least one writable container, eg one. By limiting the number of writable containers, the data written in the storage system is distributed locally on the storage device. When the writable container is full, the full writable container becomes a read-only container, and the storage system creates a new writable container for use by write requests. Writable containers are used only in response to write requests, while read-only containers are used only in response to read requests.

Referring to Figure 15, when the storage system receives a write request, it writes data to the NVDIMM. After data is written to the NVDIMM, a message is issued indicating that the write request is complete. Data is also written to writable container 3 after a write request is received. The data is written to the writable container 3 in an append-write manner, and the data that has been written to the writable container 3 will not be updated. After container 3 has been written, set container 3 as a read-only container and is no longer used to respond to write requests. Through this writing method, the write operations on the disk device in the storage system are basically sequential writing of large blocks of data, so that the FTL (Flash Translation Layer) of the disk device works in the best state, and the disk device does not Garbage collection is started frequently for maximum performance. Through this data distribution method, writing data always occurs locally in the disk device, and the disk device is operated in the form of large data blocks. Therefore, the write amplification factor within the disk device will be greatly reduced. Thereby, the service life of the disk device can be improved as a whole.

Continuing to refer to Figure 15, a read request in the storage system is responded to by a read-only container. In Figure 15, container 0, container 1 and container 2 are read-only containers. When the storage system receives a read request, it obtains the read-only container where the data corresponding to the read request is located through the mapping mechanism, and reads the data from the read-only container. After the writable container 3 is full, it also becomes a read-only container, and is no longer used for writing requests but only used for responding to read requests. When writing data into the container 3, the corresponding relationship between the data and the container 3 is recorded. After container 3 is full, container 3 becomes a read-only container.

In this way, read and write operations are separated at the container level, thereby reducing the coupling between read and write operations, reducing the interference between read and write operations, and enhancing the consistency of read and write delays. This is completely different from the traditional LBA-based data layout. In an embodiment according to the present invention, when a lot of data is written in the storage system, the written data will be distributed to almost any position of the multiple disks. At this time, when responding to a read request, data is read from multiple locations on multiple disks, so that the advantage of the high random read performance of the SSD can be fully utilized. At a certain moment, the write request occurs on the writable container, so that the target locations of the write request and the read request are separated from each other, thereby reducing the impact of the read request and the write request.

In a further embodiment, a cache is provided in the storage system for writable containers. Therefore, when the writable container such as container 3 is not yet full, the cache responds to the read request for the data written in the writable container 3 . In the embodiment according to FIG. 15, the NVDIMM may be used as a cache for the writable container for responding to read requests for data written in the writable container. In another example, the memory or other high-speed storage medium of the storage system is used as a cache for the writable container.

FIG. 16 is a flowchart of a data access method of a storage system according to an embodiment of the present invention. In response to receiving the write request (1610), data is written to a writable storage object (eg, see Figure 15, container 3) (1620). Writes data to a writable storage object in append-write mode. When the writable storage object is full (1630), the storage object is set as a read-only storage object (1640). In response to receiving the read request (1650), data is read from the read-only storage object (1660).

When data is written into the writable storage object, the corresponding relationship between the data and the storage object is also recorded. When responding to the read request, the read-only storage object storing the data is determined according to the corresponding relationship of the records, and the data is read from the read-only storage object.

FIG. 17 is a flowchart of a data access method for a storage system according to yet another embodiment of the present invention. Referring also to Figure 15, in response to receiving a write request (1710), data is written to the NVDIMM (1770). NVDIMM acts as a write cache for the storage system. After data is written to the NVDIMM, a message is sent indicating that the write request is complete (1780). This reduces the latency of write operations. In response to receiving the write request, the data is also written to the writable storage object (eg, container 3 in Figure 15) (1720). When data is written into the writable storage object, the corresponding relationship between the data and the storage object is also recorded, so as to be used for reading the data. When the writable storage object is full (1730), the writable storage object is set as a read-only (1740) storage object (1740). In one example, there is only one writable storage object in the storage system at any one time. When the writable storage object is full, the writable storage object is set to read-only. The storage system creates new writable storage objects to carry write requests. In another example, several writable storage objects exist simultaneously in the storage system. In response to receiving the read request (1750), data is read from the read-only storage object (1760).

In one example, the NVDIMM also acts as a cache for writable storage objects. If data has been written to the writable storage object, and the writable storage object is not full, a read request for the written data occurs, and the requested data is read from the NVDIMM. After the writable storage object is full, the data corresponding to the writable storage object cached in the NVDIMM is also released.

In another example, the NVDIMM acts as a write cache for the storage system. After the data is written to the writable storage object, the corresponding data is released from the NVDIMM. A storage system provides a cache of writable storage objects in memory or other high-speed storage media. In response to receiving a write request, cache data in the cache provided for writable storage objects. If data has been written to the writable storage object, and the writable storage object is not full, and a read request for the written data occurs, the requested data is read from the cache provided for the writable storage object. After the writable storage object is full, the data corresponding to the writable storage object stored in the cache provided for the writable storage object is also released.

FIG. 18 is a flowchart of a data access method of a storage system according to still another embodiment of the present invention. In response to receiving the first data write request (1810), in an embodiment according to Figure 18, the first data is written to the NVDIMM (1812). The first data write request includes first data and a first logical address to which the first data is to be written. After the first data is written to the NVDIMM, a message is sent to the application, other program, or server that issued the write request to indicate that the first data write request has been processed (1814).

In the embodiment according to FIG. 7 , after receiving the first data write request, the first data corresponding to the first write request is not written to the writable storage object immediately. Instead, wait for a second write request to be received (1820). By combining the first write request and the second write request, the number of write operations performed by the storage device can be reduced on the premise of ensuring data reliability, thereby improving the performance of the storage system.

In response to receiving the second data write request (1820), second data is written to the NVDIMM (1822). After the second data is written to the NVDIMM, a message is sent to the application, other program or server that issued the write request to indicate that the second data write request has been processed (1824).

In response to receiving the first write request and the second write request, the first data and the second data are written to a writable storage object (eg, see Figure 15, container 3) (1830). In one example, the first data is aggregated with the second data such that the blocks of data written to the writable storage object have a larger size. In another example, when the first data write request and the second data write request are consecutive in logical addresses, the second data is appended to the first data, and the writable storage object is written. In still another example, after the first data and the second data are aggregated in the NVDIMM, the aggregated data in the NVDIMM is written to the writable storage object.

Writes data to a writable storage object in append-write mode. The correspondence between the first data and the writable storage object, and the correspondence between the second data and the writable storage object are recorded. When the writable storage object is full (185), the storage object is set as a read-only storage object (1860). In response to receiving the read request (1870), the storage object storing the requested data is determined according to the correspondence between the recorded data and the storage object, and data is read from the read-only storage object (1875).

In one example, in response to writing the first and second data to the writable storage object, the first and second data are released in the NVDIMM (1840). In another example, NVDIMMs are used as a cache for writable storage objects. If data has been written to the writable storage object, and the writable storage object is not full, a read request for the written data occurs, and the requested data is read from the NVDIMM. After the writable storage object is full, the data corresponding to the writable storage object cached in the NVDIMM is also released.

It should be noted that the data merged in step 1830 may come from two or more data write requests. In another example, after receiving the second data write request, if the first data and the second data are not suitable for combining, the first data and the second data are written into the writable storage object respectively.

Embodiments of the present invention also provide a computer program comprising computer program code, which, when loaded into a computer system and executed on the computer system, causes the computer system to perform the above-described method.

Embodiments of the present invention further provide a program including program codes, when loaded into a storage device and executed on the storage device, the program code causes the storage device to execute the above-described method.

It will be understood that each block of the block diagrams and flowchart illustrations, and combinations of blocks in the block diagrams and flowchart illustrations, respectively, can be implemented by various means including computer program instructions. These computer program instructions can be loaded on a general purpose computer, special purpose computer or other programmable data control device to produce a machine such that the instructions executed on the computer or other programmable data control device create a flow diagram for implementing one or more blocks device with the function specified in .

These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data control device to function in a particular manner, such that the instructions stored in the computer readable memory can be used to manufacture including for implementing a Articles of manufacture of computer readable instructions for the functions specified in the flowchart block or blocks. Computer program instructions can also be loaded onto a computer or other programmable data control device to cause a series of operations to be performed on the computer or other programmable data control device, resulting in a computer-implemented process, which in turn is performed on the computer or other programmable data control device. The instructions executing on the control device provide operations for implementing the functions specified in one or more of the flowchart blocks.

Accordingly, blocks of the block diagrams and flowchart illustrations support combinations of means for performing the specified functions, combinations of operations for performing the specified functions and combinations of program instruction means for performing the specified functions. It will also be understood that each block of the block diagrams and flowchart illustrations, and combinations of blocks in the block diagrams and flowchart illustrations, can be implemented by special purpose hardware-based computer systems that perform the specified functions or operations, or by combinations of special purpose hardware and computer instructions.

The data access method and device of the storage system have been disclosed above. Those skilled in the art will also appreciate that the methods or operational procedures disclosed in this disclosure may be implemented by software, firmware, and any combination thereof. The software and firmware for implementing the methods or operation procedures of the embodiments of the present invention may be executed by the CPU of the host accessing the storage device. Software and firmware for implementing the methods or operations of the embodiments of the present invention may be stored in a network server, a host accessing a storage device, and/or a storage device.

Although the examples with reference to the present invention are described for purposes of explanation only and not limitation of the invention, changes, additions and/or deletions to the embodiments may be made without departing from the scope of the invention.

Many modifications and other embodiments of the inventions set forth herein will come to mind to those skilled in the art to which these embodiments pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the inventions are not to be limited to the specific embodiments disclosed, but that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims (10)

1. A data access method for a storage system, the storage system comprising a plurality of storage devices and NVDIMMs, the storage system providing a plurality of storage objects, the storage objects being composed of storage resources on the storage device, the plurality of storage The storage objects include one or more writable storage objects and multiple read-only storage objects; the method includes: in response to a write request, writing data to the NVDIMM and appending writing data to the writable storage object; In response to the completion of the operation of writing data to the NVDIMM, sending a message indicating that the write request is completed; in response to writing data to the writable storage object, releasing the space occupied by the data in the NVDIMM; If the writable storage object is full, set the writable storage object as a read-only storage object; In response to a read request, data is read from the read-only storage object. 2. The method of claim 1, wherein the storage object comprises a first storage space from a first storage device and a second storage space from a second storage device. 3. The method according to claim 2, wherein the second storage space is used to store check data of the data of the first storage space. 4. The method according to one of claims 1 to 3, further comprising recording the mapping relationship between the written data and the writable storage object. 5. The method according to claim 4, further comprising, in response to a read request, searching for a read-only storage object that stores the requested data according to the mapping relationship, and from the read-only storage object that stores the requested data Read data. 6. The method of one of claims 1-3, further comprising: The data is read from the NVDIMM in response to a read request prior to setting the writable storage object as a read-only storage object. 7. The method of one of claims 1-3, wherein the storage system further provides a cache, the method further comprising: In response to the write request, write data to the cache; The data is read from the cache in response to a read request before the writable storage object is set as a read-only storage object. 8. A data access method for a storage system, wherein the storage system includes a plurality of storage devices and NVDIMMs, the storage system provides a plurality of storage objects, and the storage objects are composed of storage resources on the storage device, and the plurality of storage objects are Each storage object includes one or more writable storage objects and multiple read-only storage objects; the method includes: writing first data to the NVDIMM in response to a first write request; In response to writing the first data to the NVDIMM, sending a message indicating that the first write request is complete; writing second data to the NVDIMM in response to a second write request; In response to writing the second data to the NVDIMM, sending a message indicating that the second write request is complete; generating a storage data block, the storage data block includes the first data and the second data; writing the storage data block to the writable storage object in an append write/sequential write (append) manner; If the writable storage object is full, set the writable storage object as a read-only storage object; The first data or the second data is read from the read-only storage object in response to a read request. 9. A computer comprising: a machine-readable memory for storing program instructions; one or more processors for executing program instructions stored in said memory; said program instructions for causing said one or A plurality of processors perform the method according to one of claims 1-8. 10. A data access device for a storage system, the storage system comprising a plurality of storage devices and NVDIMMs, the storage system providing a plurality of storage objects, and the storage objects are composed of storage resources on the storage device, the plurality of The storage objects include one or more writable storage objects and multiple read-only storage objects; the apparatus includes: a writing module, configured to write data to the NVDIMM in response to a write request, and write data to the writable storage object in an additional write manner; means for sending a message indicating the completion of the write request in response to the completion of the operation of writing data to the NVDIMM; means for releasing the space occupied by the data in the NVDIMM in response to writing data to the writable storage object; a storage object setting module, configured to set the writable storage object as a read-only storage object if the writable storage object is full; A readout module, configured to read data from the read-only storage object in response to a read request.

Images