CN109840217B - Cache resource allocation and device - Google Patents

Abstract

A method and a device for allocating cache resources comprise the steps of receiving a data writing request, wherein the data writing request is used for writing data to be written into a target volume, the storage device comprises a plurality of volumes and a cache, the space of the cache comprises a private space and a public space, the private space is divided into a plurality of parts, each part is allocated to one of the volumes as a quota, and the target volume is one of the volumes. And judging whether the debt information of the target volume is larger than zero or not, wherein the debt information is used for indicating whether the quota distributed for the target volume is overdraft or not. And if the debt information of the target volume is greater than zero, allocating cache resources for the data writing request from the public space. The efficiency of processing the service by the storage device can be improved.

Description

A cache resource allocation and device

Technical field

This application relates to the field of storage technology, in particular to a cache resource allocation and device.

Background technique

The storage device will apply for cache resources when processing business requests. The usual approach is to regard the cache space as a cache resource pool, and allocate cache resources to business requests from the cache resource pool on a first-come, first-served basis. However, services with a slow resource release speed will suppress those services with a fast resource release speed at the resource level, affecting the efficiency of the storage device in processing services.

Contents of the invention

The cache resource allocation method and device provided by this application can improve the efficiency of storage equipment processing services.

The first aspect provides a cache resource allocation method. In this method, a processor receives a write data request, which is used to write data to be written into a target volume. The storage device includes multiple volumes and a cache. The cache space includes a private space and a cache. Public space, the private space is divided into several parts, each part is allocated as a quota to one of the plurality of volumes, and the target volume is one of the plurality of volumes. The processor then determines whether the debt information of the target volume is greater than zero, and the debt information is used to indicate whether the quota allocated for the target volume has been overdrawn. If the debt information of the target volume is greater than zero, cache resources are allocated from the public space for the write data request.

According to the first aspect, the cached space includes a private space and a public space, the private space is divided into several parts, and each part is allocated as a quota to one of the plurality of volumes. When receiving a write data request, it determines how to allocate cache resources for the write data request based on the debt information of the target volume to be accessed. If the debt information of the target volume is greater than zero, cache resources are allocated from the public space for the write data request. Debt information greater than zero means that other volumes still owe some quota to the target volume. Therefore, for the write data request, cache resources can be allocated from the public space to avoid the target volume's quota being further occupied and affecting efficiency. This improves the processing efficiency of write data requests for the target volume.

With reference to the first aspect, in a first implementation manner of the first aspect, when the cache resources allocated for the write data request are released, the processor adds the released cache resources to the cache resources allocated for the target volume. of quota. According to the first implementation, when the debt information is greater than 0, it means that other volumes owe the target volume some quotas, and the released cache resources are added to the quota allocated for the target volume to avoid other volumes owing more quotas to the target volume. This drags down the processing efficiency of write data requests for the target volume.

In conjunction with the first aspect, in a second implementation manner of the first aspect, the debt information is equal to the updated value of the baseline quota after applying for or releasing cache resources in the current cycle, and the baseline quota is equal to the new quota minus the old quota. The difference between the quotas, the new quota is the quota allocated to the target volume in the current period, and the old quota is the quota allocated to the target volume in the historical period.

With reference to the first aspect, in a third implementation manner of the first aspect, each volume corresponds to a buffer queue, and the buffer queue stores a write data request or is waiting to be processed. The processor allocates the quota to the volume according to the number of write data requests contained in the buffer queue corresponding to each volume and the size of the write data requests. The greater the number of write data requests contained in the buffer queue, the higher the number of write data requests contained in the buffer queue. The more quotas are allocated to the corresponding volume in the queue and the larger the size of the write data request contained in the buffer queue, the more quotas are allocated to the corresponding volume in the buffer queue. According to the third implementation manner, a quota is allocated to each volume according to its business pressure, and cache resources are tilted toward volumes with greater business pressure, thereby improving the efficiency of the storage device in serving services.

In connection with the first aspect, in a fourth implementation manner of the first aspect, the processor allocates the quota to the volume according to the cache concurrency number of each volume, the cache concurrency number is used to indicate when the data in the cache is copied to the hard disk. The number of write requests that can be written to the volume concurrently. According to the fourth implementation manner, a quota is allocated to each volume according to its corresponding cache concurrency number, and cache resources are tilted toward volumes with high concurrency capabilities, thereby improving the efficiency of storage device service services.

With reference to the second implementation manner of the first aspect, in a third implementation manner of the first aspect, the historical period is a period before the current previous period.

A second aspect of this application provides a cache resource allocation device, which is configured to execute the first aspect and the method in any implementation of the first aspect.

A third aspect of this application provides a storage device, including a processor and a cache. The processor calls the program code in the cache to implement the method in the first aspect and any implementation manner of the first aspect.

A fourth aspect of the present application provides a computer program product, including a computer-readable storage medium storing program code, and the program code includes instructions that can be implemented by at least one method of the first aspect.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present invention more clearly, the drawings required to be used in the embodiments will be briefly introduced below.

Figure 1 is a system architecture diagram provided by an embodiment of the present invention;

Figure 2 is a schematic diagram of the cache space provided by an embodiment of the present invention;

Figure 3 is a schematic flowchart of a cache resource allocation method provided by an embodiment of the present invention;

Figure 4 is a schematic flowchart of a cache resource release method provided by an embodiment of the present invention;

Figure 5 is a schematic structural diagram of a cache resource allocation device provided by an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

Figure 1 is a system architecture diagram provided by an embodiment of the present invention. The storage system provided by this embodiment includes a host 20, a controller 11 and multiple hard disks 22. Controller 11 may be a computing device such as a server, desktop computer, etc. An operating system and application programs are installed on the controller 11 . The controller is connected to the host 20 through a storage area network (SAN). Specifically, the host 20 may send a write data request to the controller 11. After receiving the write data request, the controller 11 allocates cache resources to the controller 11 to temporarily store the data carried in the write data request in the cache 102. When the data stored in the cache 102 reaches a certain level, the data in the cache 102 is written to the hard disk 22 . In addition, the host 20 can also send a read data request to the controller 11. After receiving the read data request, the controller 11 searches whether there is data to be read in its cache 102 according to the address in the read data request. If so, Then the data to be read is directly sent to the host 20. If not, the data is obtained from the hard disk 22 and sent to the host 20. In practical applications, the controller 11 and the hard disk 22 may be integrated in one storage device, or may be located in two independent devices. The embodiment of the present invention places no restrictions on the positional relationship between the controller 11 and the hard disk 22 . For convenience of description, in this embodiment, the controller 11 and the hard disk 22 are collectively referred to as a storage device.

In actual applications, the hard disks 22 of the storage devices are logically combined and the required RAID levels are applied to obtain a RAID set. Because a RAID set combines multiple hard disks and usually has a large capacity, the available capacity of the RAID set is divided into smaller units, called volumes. The volumes hide the organization and composition of the RAID set from the host 20 . A volume can be allocated to one host 20 for use during initialization, or can be allocated to multiple hosts 20 for use.

The structure of the controller 11 is introduced below. As shown in Figure 1, the controller 11 at least includes a processor 101, a cache 102 and an interface (not shown in Figure 1). The processor 101 is a central processing unit (English: central processing unit, CPU). In the embodiment of the present invention, the processor 101 may be configured to receive read data requests and write data requests from the host 20 and process the read data requests and write data requests.

The cache 102 is used to temporarily store data received from the host 20 or data read from the hard disk 22 . When the controller 11 receives multiple write data requests sent by the host, the data in the multiple write data requests may be temporarily stored in the cache 102 . When the capacity of the cache 102 reaches a certain threshold, the data stored in the cache 102 is sent to the hard disk 22 . Hard disk 22 stores the data. Cache 102 includes volatile memory, non-volatile memory, or a combination thereof. Volatile memory is, for example, random-access memory (English: random-access memory, RAM). Non-volatile memories include floppy disks, hard disks, solid state disks (SSD), optical disks, and other machine-readable media that can store program codes. It can be understood that the storage device needs to use the cache 102 when executing a write data request or a data read request.

As shown in Figure 2, the storage space of cache 102 includes private space and public space. The private space can be divided into several parts, and each part is allocated to a volume as a quota. For example, quota 1 is assigned to volume 1, quota 2 is assigned to volume 2, and quota n is assigned to volume n. The quota allocated to each volume can be the same or different. When performing initial settings, the controller 11 may allocate the private space to each volume equally, or may allocate it according to certain principles. After the initial allocation, every cycle, the processor 101 will adjust the quota allocated to each volume according to the front-end business pressure or the back-end data flushing capability to obtain a new quota. The cycle refers to the same time interval, and in the embodiment of the present invention, the cycle management can be implemented by a timer. The new quota is relative to the old quota. Specifically, the new quota refers to the quota allocated in the current cycle, and the old quota refers to the quota allocated in the historical cycle. Common space is not pre-allocated to any volume, each volume can use said common space based on its own debt information.

In one case, the processor 101 can adjust the quota of each volume according to the front-end business pressure. Each volume corresponds to a buffer queue, and the buffer queue stores a write data request or is waiting to be processed. The quota is allocated to the volume according to the number of write data requests contained in the buffer queue corresponding to each volume and the size of the write data requests. The greater the number of write data requests contained in the buffer queue, the larger the number of write data requests contained in the buffer queue. The more quotas are allocated to the corresponding volume, the larger the size of the write data request contained in the buffer queue, and the more quotas are allocated to the corresponding volume in the buffer queue. Specifically, the product of the number of queued write data requests corresponding to each volume and the preset size of the write data requests can be calculated. The larger the product, the more quotas are allocated. The cache queue is located in cache 102.

In another case, the processor 101 can adjust the quota of each volume according to the backend data flushing capability. Specifically, the quota may be allocated to the volume according to the cache concurrency number of each volume. The cache concurrency number is used to indicate the number of write requests that can be written to the volume concurrently when the data in the cache is copied to the hard disk. The write request here refers to a request for writing data from the cache 102 to the hard disk 22 , which is different from a write data request received from the host for writing to the controller 11 .

Please refer to Figure 3 below. Figure 3 shows a cache resource allocation method provided in this embodiment. This method can be applied in the system shown in Figure 1 and is executed by the processor 101. Specifically, the method includes the following steps.

In S301, the processor 101 receives a write data request, which includes an address of data to be written. The volume to which the data to be written can be determined based on this address. For convenience of description, the volume to which the data to be written is called the target volume.

In S302, the processor 101 determines whether the debt information of the target volume is greater than, less than, or equal to 0. The debt information is used to indicate whether the quota allocated for the target volume has been overdrawn. If the debt information is greater than 0, it means that other volumes owe some quota to the target volume; if the debt information is equal to 0, it means that the quota of the target volume is just enough, and it does not owe other volumes, and other volumes do not owe the target volume. If the debt information is less than 0, it means that the target volume owes some quota to other volumes.

Specifically, the debt information is equal to the updated value of the baseline quota after applying for or releasing cache resources in the current cycle. The baseline quota is equal to the difference obtained by subtracting the old quota from the new quota. The new quota is the quota allocated to the target volume in the current period. The old quota is the quota allocated to the target volume in the historical period. The historical period may be the period before the current period, or any one of several previous periods. For example, assuming that the new quota of the target volume is 2000 and the old quota of the previous period is 1000, then the baseline quota is equal to 1000. The base quota is greater than 0, indicating that other volumes owe the quota of the target volume 1000 at this time. In the current cycle, if the processor 101 processes two data write requests for the target volume and applies for a quota of 200 and a quota of 300 respectively, then the debt information at this time is equal to 1000-200-300=500. In S302, if the debt information of the target volume is greater than 0, execute S303; if the debt information of the target volume is equal to 0, execute S304; if the debt information of the target volume is less than 0, execute S303.

In S303, since the debt information is greater than 0, it means that other volumes owe some quota to the target volume, so the cache resources allocated for the write data request at this time can be allocated from the public space.

In S304, since the debt information is equal to 0, it means that the quota of the target volume is just enough, so the cache resources allocated for the write data request at this time still use the quota allocated for the target volume.

In S305, since the debt information is less than 0, it means that the target volume owes some quotas to other volumes, so the cache resources allocated for the write data request at this time still use the quota allocated for the target volume.

According to the implementation shown in Figure 3, the cached space includes private space and public space. The private space is divided into several parts, and each part is allocated as a quota to one of the multiple volumes. When receiving a write data request, it determines how to allocate cache resources for the write data request based on the debt information of the target volume to be accessed. If the debt information of the target volume is greater than zero, cache resources are allocated from the public space for the write data request. Debt information greater than zero means that other volumes still owe some quota to the target volume. Therefore, for the write data request, cache resources can be allocated from the public space to avoid the target volume's quota being further occupied and affecting efficiency. This improves the processing efficiency of write data requests for the target volume.

It can be understood that when data is deleted, the cache resources occupied by it can be released. This embodiment determines whether to add the released cache resources to the quota allocated for the volume or to the public space based on the debt information of each volume. Please refer to Figure 4, which is a schematic flowchart of cache resource release according to an embodiment of the present invention. As shown in Figure 4, it includes the following steps:

In S401, the processor 101 receives a data deletion request, which includes the address of the data. The volume to which the data to be written can be determined based on this address. Here we still take the target volume as an example.

For S402, reference may be made to the description of S302 in Figure 3, which will not be described again here.

In S403, since the debt information is greater than 0, it means that other volumes owe some quota to the target volume, so the cache resources released at this time should be returned to the target volume. Accordingly, the operation is to add the released cache resources to the target volume. of quota.

In S404, since the debt information is equal to 0, it means that the quota of the target volume is just enough, so the cache resources allocated for the write data request at this time are still returned to the target volume.

In S405, since the debt information is less than 0, it means that the target volume owes some quotas to other volumes. Therefore, the cache resources allocated for write data requests at this time should be returned to other volumes. Correspondingly, the operation is to add the released cache resources to in public space.

According to the implementation shown in Figure 4, when the debt information is greater than 0, it means that other volumes owe the target volume some quotas, and the released cache resources are added to the quota allocated for the target volume to avoid other volumes owing quotas to the target volume. More and drag down the processing efficiency of write data requests for the target volume.

Please refer to Figure 5. Figure 5 is a cache resource allocation device 50 provided by an embodiment of the present invention. The device 50 is located in a storage device and includes a receiving module 501, a judgment module 502, and an allocation module 503.

The receiving module 501 is configured to receive a write data request. The write data request is used to write the data to be written into the target volume. The storage device includes multiple volumes and caches. The cache space includes private space and public space. space, the private space is divided into several parts, each part is allocated as a quota to one of the plurality of volumes, and the target volume is one of the plurality of volumes. The receiving module 501 can be implemented by the processor 101 calling the program code in the cache 102. For its specific implementation, refer to S301 shown in Figure 3.

The judgment module 502 is used to judge whether the debt information of the target volume is greater than zero, and the debt information is used to indicate whether the quota allocated for the target volume has been overdrawn. The judgment module 502 can be implemented by the processor 101 calling the program code in the cache 102. For its specific implementation, reference can be made to S302 shown in Figure 3.

The allocation module 503 is configured to allocate cache resources from the public space to the write data request if the debt information of the target volume is greater than zero. The allocation module 503 can be implemented by the processor 101 calling the program code in the cache 102. For its specific implementation, refer to S303 shown in Figure 3.

Optionally, the apparatus 50 may further include a release module 504, configured to add the released cache resources to the quota allocated for the target volume when the cache resources allocated for the write data request are released. The release module 504 can be implemented by the processor 101 calling the program code in the cache 102. For its specific implementation, refer to S403 shown in Figure 4.

Persons of ordinary skill in the art will understand that various aspects of the present invention, or possible implementations of various aspects, may be embodied as systems, methods, or computer program products. Accordingly, aspects of the invention, or possible implementations of aspects, may take the form of entirely hardware embodiments, entirely software embodiments (including firmware, resident software, etc.), or embodiments combining software and hardware aspects, in Here they are collectively referred to as "circuit", "module" or "system". In addition, various aspects of the present invention, or possible implementations of various aspects, may be in the form of a computer program product. The computer program product refers to a computer-readable program code stored in a computer-readable medium.

Computer-readable media includes, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any appropriate combination of the foregoing, such as random access memory (RAM), read only memory (ROM), removable Programmable read-only memory (EPROM), optical disk.

The processor in the computer reads the computer-readable program code stored in the computer-readable medium, so that the processor can perform the functional actions specified in each step or a combination of steps in the flowchart.

The computer-readable program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server . It should also be noted that in some alternative implementations, the steps in the flowchart illustrations, or the functions noted in blocks in the block diagrams, may occur out of the order noted in the figures. For example, two steps, or blocks shown in succession may in fact be executed approximately concurrently, or the blocks may sometimes be executed in the reverse order, depending on the functionality involved.

Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. One of ordinary skill in the art may implement the described functionality using different methods for each specific application, but such implementations should not be considered to be beyond the scope of the present invention.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Those of ordinary skill in the art can easily think of changes or replacements within the technical scope disclosed by the present invention, and all of them should be covered. within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (10)

1. A cache resource allocation method, characterized by including: Receive a write data request. The write data request is used to write the data to be written into the target volume. The storage device includes multiple volumes and caches. The cache space includes private space and public space. The private space is divided Be a plurality of parts, each part being allocated as a quota to one of the plurality of volumes, and the target volume is one of the plurality of volumes; Determine whether the debt information of the target volume is greater than zero, and the debt information is used to indicate whether the quota allocated for the target volume has been overdrawn; If the debt information of the target volume is greater than zero, allocate cache resources for the write data request from the public space; When the cache resources allocated for the write data request are released, the released cache resources are added to the quota allocated for the target volume. 2. The method according to claim 1, characterized in that the debt information is equal to the value after the baseline quota is updated after applying for cache resources in the current cycle, or the debt information is equal to the value after the cache resources are released in the current cycle. The updated value of the baseline quota. The baseline quota is equal to the difference obtained by subtracting the old quota from the new quota. The new quota is the quota allocated to the target volume in the current cycle. The old quota is the quota allocated to the target volume in the historical cycle. The allocated quota for the target volume. 3. The method according to claim 1, characterized in that each volume corresponds to a buffer queue, and the buffer queue stores a write data request or is waiting to be processed, and further includes: The quota is allocated to the volume according to the number of write data requests contained in the buffer queue corresponding to each volume and the size of the write data requests. The greater the number of write data requests contained in the buffer queue, the higher the number of write data requests contained in the buffer queue. The more quotas are allocated to a volume and the larger the size of the write data request contained in the buffer queue, the more quotas are allocated to the corresponding volume in the buffer queue. 4. The method of claim 1, further comprising: The quota is allocated to the volume according to the cache concurrency number of each volume, which is used to indicate the number of write requests that can be written to the volume concurrently when the data in the cache is copied to the hard disk. 5. The method according to claim 2, characterized in that the historical period is a period before the current period. 6. A cache resource allocation device, characterized in that the device is located in a storage device and includes: A receiving module, configured to receive a write data request. The write data request is used to write the data to be written into the target volume. The storage device includes multiple volumes and caches. The cache space includes private space and public space. , the private space is divided into several parts, each part is allocated as a quota to one of the plurality of volumes, and the target volume is one of the plurality of volumes; A judgment module, used to judge whether the debt information of the target volume is greater than zero, and the debt information is used to indicate whether the quota allocated for the target volume has been overdrawn; An allocation module, configured to allocate cache resources from the public space to the write data request if the debt information of the target volume is greater than zero; A release module, configured to add the released cache resources to the quota allocated for the target volume when the cache resources allocated for the write data request are released. 7. The device according to claim 6, wherein the debt information is equal to the updated value of the baseline quota after applying for cache resources in the current cycle, or the debt information is equal to the value after releasing the cache resources in the current cycle. The updated value of the baseline quota. The baseline quota is equal to the difference obtained by subtracting the old quota from the new quota. The new quota is the quota allocated to the target volume in the current cycle. The old quota is the quota allocated to the target volume in the historical cycle. The allocated quota for the target volume. 8. The device according to claim 6, wherein each volume corresponds to a buffer queue, and a write data request is stored in the buffer queue or is waiting to be processed, The allocation module is also configured to allocate the quota to the volume according to the number of write data requests contained in the buffer queue corresponding to each volume and the size of the write data requests, wherein the number of write data requests contained in the buffer queue is The larger the number, the more quotas are allocated to the volume corresponding to the buffer queue, and the larger the size of the write data request contained in the buffer queue, the more quotas are allocated to the corresponding volume of the buffer queue. 9. The device according to claim 6, characterized in that, The allocation module is also configured to allocate the quota to the volume according to the cache concurrency number of each volume. The cache concurrency number is used to indicate the number of write requests that can be written to the volume concurrently when the data in the cache is copied to the hard disk. quantity. 10. The device according to claim 7, wherein the historical period is a period before the current period.