CN103870312B - Establish the method and device that virtual machine shares memory buffers - Google Patents

Abstract

The invention discloses a kind of method and device established virtual machine and share memory buffers, this method includes：Obtain in virtual machine start-up course from the virtual data for starting disk read-write；The data are stored to virtual machine shared buffer memory device, the data are read from the virtual machine shared buffer memory device when starting for other virtual machines.By the above-mentioned means, the present invention can only be buffered in effective data during virtual machine starts with seldom memory headroom, solve the problems, such as input and output when virtual machine concurrently starts read-write data.

Description

Method and device for establishing virtual machine shared storage cache

technical field

The invention relates to the technical field of virtual machines, in particular to a method and device for establishing a virtual machine shared storage cache.

Background technique

Regardless of whether it is a virtual machine deployed in full-clone mode or a virtual machine deployed in linked-clone mode, when a large number of virtual machines are started at the same time, frequent input and output (IO, Input Output) operations occur, which creates a negative impact on shared storage. The high input and output per second (IOPS, Input Output per second) pressure peak causes the virtual machine to start slowly. When configuring desktop cloud storage, configure it according to the highest peak requirement, resulting in waste of high IOPS storage configuration during normal hours.

The existing technology adopts the flow control method, sets the start time of the virtual machine several hours before going to work, and gradually starts all the virtual machines in advance in a flow control manner, so as to avoid the impact of centralized concurrent start on the storage service capability.

The prior art does not really solve the input/output problem when the virtual machine starts to read and write data concurrently.

Contents of the invention

The technical problem mainly solved by the present invention is to provide a method and device for establishing a virtual machine shared storage cache, which can only cache valid data during the startup of the virtual machine with a small memory space, and solve the problem of reading and writing data when the virtual machine starts concurrently. Input and output problems.

The first aspect of the present invention provides a method for establishing a virtual machine shared storage cache, including: obtaining data read and written from the virtual boot disk during the virtual machine startup process; storing the data to the virtual machine shared cache device for other virtual machines The data is read from the virtual machine shared cache device when the machine starts.

In a first possible implementation manner of the first aspect, before the step of acquiring the data read and written from the virtual boot disk during the startup process of the virtual machine, the step includes: the virtual machine sends the read data to the virtual machine shared cache device A write request: the virtual machine receives a read and write request response sent by the virtual machine shared cache device, and the read and write request response carries read and write request failure information.

With reference to the first aspect or the first possible implementation of the first aspect, in the second possible implementation of the first aspect, the method further includes: repeating the data stored in the virtual machine shared cache device Data deletion operations.

With reference to the first aspect or the first possible implementation of the first aspect, in a third possible implementation of the first aspect, the virtual machine startup process includes booting the virtual machine into the virtual machine operating system The startup is completed or the virtual machine is powered on and the startup of the preset application program of the virtual machine is completed.

The second aspect of the present invention provides a virtual machine shared cache device, the device includes: an acquisition module and a storage module; the acquisition module is used to acquire data read and written from the virtual boot disk during the startup process of the virtual machine; the The storage module is used for storing the data in the virtual machine shared cache device, so that other virtual machines can read the data from the virtual machine shared cache device when starting.

In a first possible implementation manner of the second aspect, the device further includes: a read-write module; the read-write module is configured to, when the virtual machine sends a read-write request to the virtual machine shared cache device, receiving the request, and sending a read and write request response sent by the virtual machine shared cache device to the virtual machine, where the read and write request response carries read and write request failure information.

With reference to the second aspect or the first possible implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the device further includes a deduplication module, and the deduplication module is configured to perform the The virtual machine shares data stored in the cache device to perform a data deduplication operation.

With reference to the second aspect or the first possible implementation of the second aspect, in a third possible implementation of the second aspect, the virtual machine startup process includes booting the virtual machine into the virtual machine operating system The startup is completed or the virtual machine is powered on and the startup of the preset application program of the virtual machine is completed.

The beneficial effects of the present invention are: different from the situation of the prior art, the present invention obtains the data read and written from the virtual boot disk during the virtual machine startup process; only caches the data valid for the virtual machine startup to the virtual machine shared cache device, so as to It is used to read data from the virtual machine shared cache device when other virtual machines are started. In this way, only valid data during virtual machine startup can be cached with a small memory space, and the input and output problem of concurrent startup of virtual machines can be solved.

Description of drawings

FIG. 1 is a flowchart of an embodiment of a method for establishing a virtual machine shared storage cache in the present invention;

FIG. 2 is a flow chart of another embodiment of the method for establishing a virtual machine shared storage cache in the present invention;

FIG. 3 is an implementation of the method for establishing a virtual machine shared storage cache in a practical application in the present invention;

FIG. 4 is another implementation in practical application of the method for establishing a virtual machine shared storage cache in the present invention;

Fig. 5 is the process of creating, starting and shutting down a virtual machine;

FIG. 6 is an implementation of the method for establishing a virtual machine shared storage cache in a practical application to de-duplicate operations in the present invention;

FIG. 7 is a schematic structural diagram of an embodiment of a device for sharing a cache among virtual machines according to the present invention;

FIG. 8 is a schematic structural diagram of another embodiment of an apparatus for sharing a cache among virtual machines according to the present invention.

detailed description

The present invention will be described in detail below in conjunction with the accompanying drawings and embodiments.

Referring to FIG. 1, FIG. 1 is a flowchart of an embodiment of a method for establishing a virtual machine shared storage cache according to the present invention, including:

Step S101: Obtain the data read and written from the virtual boot disk during the boot process of the virtual machine.

A virtual machine refers to a complete computer system that is simulated by software and has complete hardware system functions and runs in a completely isolated environment.

When the virtual machine is started, the virtual machine reads and writes data from the virtual boot disk. These data are the real and valid data needed for the virtual machine to start, and these valid data are obtained to facilitate storage of these virtual machines. Read and write from the virtual boot disk when the virtual machine starts The data.

Step S102: Store the data in the virtual machine shared cache device, so that other virtual machines can read the data from the virtual machine shared cache device when starting up.

After obtaining the data read and written from the virtual boot disk during the virtual machine startup process, these data are stored in the virtual machine shared cache device, so that other virtual machines can read the data from the virtual machine shared cache device when starting, thus sharing Most of the virtual boot disk's I/O stress.

It should be noted that, in this embodiment, the process of starting the virtual machine includes starting the virtual machine to the completion of starting the operating system of the virtual machine or starting the virtual machine to the completion of starting the preset application program of the virtual machine. For example, it may be the startup process of the Windows operating system, or the startup process of application programs such as Office and Photoshop. In the embodiment of the present invention, virtual machine startup is the startup process of a temporary virtual machine before the large-scale startup of the virtual machine cluster. According to the startup of the temporary virtual machine, the necessary read and write data for the large-scale startup process of the virtual machine are obtained, and these read and write data are read and written. The write data is stored in the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines are started.

The present invention obtains the data read and written from the virtual boot disk during the startup process of the virtual machine; only the data valid for the startup of the virtual machine is cached to the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines start In this way, only valid data during virtual machine startup can be cached with a small amount of memory space, and the problem of input and output when virtual machines are started concurrently to read and write data can be solved.

Referring to FIG. 2, FIG. 2 is a flow chart of another embodiment of the method for establishing a virtual machine shared storage cache according to the present invention, including:

Step S201: the virtual machine sends a read and write request to the virtual machine shared cache device.

When the virtual machine is started, the virtual machine sends a read and write request to the virtual machine shared cache device.

Step S202: The virtual machine receives the read and write request response sent by the virtual machine shared cache device, and the read and write request response carries failure information of the read and write request.

When there is no data in the virtual machine shared cache device for the read and write request of the virtual machine, a read and write request response is sent to the virtual machine, and the virtual machine then receives the read and write request response sent by the virtual machine shared cache device. The read and write request response carries the read and write Request failure information.

Step S203: Obtain the data read and written from the virtual boot disk during the boot process of the virtual machine.

Step S204: Store the data in the virtual machine shared cache device, so that other virtual machines can read the data from the virtual machine shared cache device when starting up.

In practical application, please refer to FIG. 3 and FIG. 4 for the implementation process of step S201, step S202, step S203, and step S204. FIG. This implementation is the process of requesting data reads during the startup of the virtual machine and the startup of the operating system, including:

(b1) Read request: The read request of the virtual machine VM0 to the virtual disk started by the operating system arrives at the read-write module (ie, module B) of the virtual machine shared cache device.

(b2) Cache (that is, the storage module of the virtual machine shared cache device, module C) query: the module B obtains the currently requested data from the module C.

(b3.2) Cache miss: that is, there is no data currently requested by module B in module C, and module C returns empty data and cache miss status to module B.

(b4) Read storage device: Module B reads the currently requested data from the storage device (that is, the virtual boot disk of the operating system).

(b5) Return from reading the storage device: the storage device returns the currently requested data to the B module.

(b6.2) Virtual machine read request return: Module B returns the data read from the storage device to the virtual machine.

(b7) Module B caches the data read from the storage device to module C. This caching operation only occurs during virtual machine startup.

(b3.1) Cache hit: that is, the data currently requested by the B module is stored in the C module, and the C module returns the read data and the cache hit status to the B module.

(b6.1) Virtual machine read request return: B module returns the data returned by C module to the virtual machine.

FIG. 4 is another implementation of the method for establishing a virtual machine shared storage cache according to the present invention in practical applications. This implementation is the process of data write request during the startup of the virtual machine and the startup of the operating system, including:

(c1) Write request: The write request of the virtual machine VM0 to the virtual boot disk of the operating system arrives at the read-write module (ie, module B) of the virtual machine shared cache device.

(c2) Write storage device: Module B sends a write request to the storage device (ie, the virtual boot disk of the operating system).

(c3) Write storage device return: the storage device returns the completion status of the write request to module B.

(c4) Update cache: If c3 returns that the write to the storage device is successful, cache the data of this write request to the C module (that is, the storage module of the virtual machine shared cache device). This cache operation only occurs during the virtual machine startup process.

(c5) Return of the virtual machine write request: module B returns the result returned by writing the storage device to the virtual machine.

Please refer to FIG. 5 . FIG. 5 is the process of creating, starting and shutting down the virtual machine VMO. The process includes:

(a1) Create a virtual machine VM0 in a non-linked clone mode from the operating system boot disk in the Hypervisor.

(a2) Start the virtual machine VM0, and record the start time as T1.

(a3) VM0 enters the boot and operating system startup process.

(a4) VM0 completes the booting and operating system startup process, and records the time when the OS startup is completed as T2.

(a5) Log in to VM0, VM0 starts to log in, and record the time when logging in to the OS is completed as T3.

In FIG. 5 , the identification of time T1 may be the time when the management software orders the virtual machine to start.

Taking the virtual machine based on the Xen system as an example, the T2 time can be identified by the following method: implement and deploy an Agent program in the virtual machine operating system in advance, and the Agent program monitors the opening and closing process of the system process. When the Agent detects that a process starts to start, it writes a mark to Xenstore. The Cache optimization module (that is, the storage module of the virtual machine shared cache device) in the Hypervisor or in the operating system where the Hypervisor is located determines the time T2 and T3 by detecting corresponding marks in the Xenstore. Taking the Windows operating system as an example, if "winlogon.exe" is detected, T2 can be determined; if "explorer.exe" is detected, T3 can be determined.

By caching data between T1 and T2 (or T3), the amount of data that needs to be cached can be effectively reduced.

Step S205: Deduplicate data stored in the virtual machine shared cache device.

Data deduplication technology is a mainstream and very popular storage technology, which can effectively optimize storage capacity. It eliminates redundant data by deleting duplicate data in the dataset and keeping only one copy. This technology can greatly reduce the demand for physical storage space, thereby meeting the growing demand for data storage. This technology can bring many practical benefits, such as: can effectively control the rapid growth of data; increase effective storage space, improve storage efficiency; save total storage cost and management cost; save network bandwidth for data transmission; save space, power supply, Cooling and other operation and maintenance costs and so on.

In practical application, please refer to FIG. 6 for the implementation process of step S205. FIG. 6 is an implementation manner of deduplication operation in practical application of the method for establishing a virtual machine shared storage cache in the present invention. This implementation is the process of deduplicating the cached data between different virtual machines. Specifically, the process of saving the data of the virtual machine VM1 to VM1_cacheMap is used as an example. The process includes:

(d1) Query whether there is a mapping entry of {start sector, length} in VM1_CacheMap, if not, create a new entry and fill in {start sector, length,-1}, and set it as the VM1_cacheMap entry to be operated; if query To the mapping entry, set the mapping entry as the VM1_cacheMap entry to be operated.

(d2) Calculate the hash value of the data to be cached.

(d3) Search the hash value in the shared cache data table. If a data item with a hash value is found, compare the data to be cached with the data in the data item bit by bit. If they are exactly the same, update the VM1_cacheMap entry currently to be operated. The Data Pointer points to the data item, and adds 1 to the ref count in the data item.

(d4) If no item with duplicate content is found in the shared cache data table, create a new data table item, fill in the corresponding content and update the DataPointer of the VM1_cacheMap table item currently to be operated to point to the newly created data item.

It should be noted that, in this embodiment, the starting process of the virtual machine includes starting the virtual machine to the completion of starting the virtual machine operating system or starting the virtual machine to the completion of starting the preset application program of the virtual machine. For example, it may be the startup of the Windows operating system, or the startup of application programs such as Office and Photoshop. In the embodiment of the present invention, virtual machine startup is the startup process of a temporary virtual machine before the large-scale startup of the virtual machine cluster. According to the startup of the temporary virtual machine, the necessary read and write data for the large-scale startup process of the virtual machine are obtained, and these read and write data are read and written. The write data is stored in the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines are started.

The present invention obtains the data read and written from the virtual boot disk during the startup process of the virtual machine; only the data valid for the startup of the virtual machine is cached to the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines start In this way, only valid data during virtual machine startup can be cached with a small memory space, and the input and output problem of concurrent startup of virtual machines can be solved. In addition, deduplication using deduplication technology saves memory.

Referring to FIG. 7 , FIG. 7 is a schematic structural diagram of an embodiment of a virtual machine shared cache device according to the present invention. The device includes: an acquisition module 101 and a storage module 102 .

Optionally, the device in FIG. 7 may execute the steps in FIG. 1 to FIG. 2 .

The obtaining module 101 is used for obtaining the data read and written from the virtual boot disk during the virtual machine startup process.

When the virtual machine is started, the virtual machine reads and writes data from the virtual boot disk. These data are the real and valid data needed for the virtual machine to start, and these valid data are obtained to facilitate storage of these virtual machines. Read and write from the virtual boot disk when the virtual machine starts The data.

The storage module 102 is used for storing the data in the virtual machine shared cache device, so that other virtual machines can read the data from the virtual machine shared cache device when starting up.

After obtaining the data read and written from the virtual boot disk during the virtual machine startup process, these data are stored in the virtual machine shared cache device, so that other virtual machines can read the data from the virtual machine shared cache device when starting, thus sharing Most of the virtual boot disk's I/O stress.

It should be noted that, in this embodiment, the process of starting the virtual machine includes starting the virtual machine to the completion of starting the operating system of the virtual machine or starting the virtual machine to the completion of starting the preset application program of the virtual machine. For example, it may be the startup of the Windows operating system, or the startup of application programs such as Office and Photoshop. In the embodiment of the present invention, virtual machine startup is the startup process of a temporary virtual machine before the large-scale startup of the virtual machine cluster. According to the startup of the temporary virtual machine, the necessary read and write data for the large-scale startup process of the virtual machine are obtained, and these read and write data are read and written. The write data is stored in the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines are started.

The present invention obtains the data read and written from the virtual boot disk during the startup process of the virtual machine; only the data valid for the startup of the virtual machine is cached to the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines start In this way, only valid data during virtual machine startup can be cached with a small amount of memory space, and the input and output problems when virtual machines are started to read and write data concurrently can be solved, and it is not limited to the deployment mode of virtual machines.

Referring to FIG. 8 , FIG. 8 is a schematic structural diagram of another embodiment of a virtual machine shared cache device according to the present invention. The device includes: an acquisition module 201 , a storage module 202 , a read-write module 203 and a deduplication module 204 .

Optionally, the device in FIG. 8 may execute the steps in FIG. 2 .

The obtaining module 201 is used for obtaining the data read and written from the virtual boot disk during the virtual machine startup process.

The storage module 202 is used for storing the data in the virtual machine shared cache device, so that other virtual machines can read the data from the virtual machine shared cache device when starting up.

The read-write module 203 is configured to receive the request when the virtual machine sends a read-write request to the virtual machine shared cache device, and send a read-write request response sent by the virtual machine shared cache device to the virtual machine, the read-write request response carrying a read-write request Request failure information.

The deduplication module 204 is configured to perform deduplication operation on the data stored in the virtual machine shared cache device.

Wherein, the starting process of the virtual machine includes starting the virtual machine to the completion of starting the operating system of the virtual machine or starting the virtual machine to the completion of starting the preset application program of the virtual machine. In the embodiment of the present invention, virtual machine startup is the startup process of a temporary virtual machine before the large-scale startup of the virtual machine cluster. According to the startup of the temporary virtual machine, the necessary read and write data for the large-scale startup process of the virtual machine are obtained, and these read and write data are read and written. The write data is stored in the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines are started.

The invention obtains the data read and written from the virtual boot disk during the startup process of the virtual machine; only the data valid for the startup of the virtual machine is cached to the virtual machine shared cache device for reading data from the virtual machine shared cache device when other virtual machines start In this way, only valid data during virtual machine startup can be cached with a small amount of memory space, and the input and output problem when virtual machines are started concurrently to read and write data can be solved. In addition, deduplication using deduplication technology saves memory.

In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device implementations described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be Incorporation may either be integrated into another system, or some features may be omitted, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, and other media that can store program codes. .

The above is only the embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the description of the present invention and the contents of the accompanying drawings, or directly or indirectly used in other related technologies fields, all of which are equally included in the scope of patent protection of the present invention.

Claims (6)

1. A kind of 1. method established virtual machine and share memory buffers, it is characterised in that including：

   According to the startup of temporary virtual machine, obtain in virtual machine start-up course from virtual startup disk read-write, described virtual machine Real effective data required for starting, the virtual machine start described in one before starting on a large scale for a virtual machine group of planes Temporary virtual machine start-up course；

   The data are stored to virtual machine shared buffer memory device, are shared when concurrently starting for other virtual machines from the virtual machine Buffer storage reads the data；

   Methods described also includes：Data de-duplication operations are carried out to the data of virtual machine shared buffer memory device storage.
2. 2. according to the method for claim 1, it is characterised in that start magnetic from virtual in the acquisition virtual machine start-up course Before the step of data of disk read-write, including：

   The virtual machine sends read-write requests to the virtual machine shared buffer memory device；

   The virtual machine receives the read-write requests response that the virtual machine shared buffer memory device is sent, and the read-write requests response is taken Band read-write requests failure information.
3. 3. method according to claim 1 or 2, it is characterised in that the virtual machine start-up course includes the virtual machine Start shooting to the VME operating system start completion or virtual machine start and arrive the default application program of the virtual machine Start completion.
4. 4. a kind of virtual machine shared buffer memory device, it is characterised in that described device includes：Acquisition module and memory module；

   The acquisition module is used for the startup according to temporary virtual machine, obtains in virtual machine start-up course and is read from the virtual disk that starts Virtual machine write, described starts required real effective data, and the virtual machine starts to be opened on a large scale for a virtual machine group of planes A temporary virtual machine start-up course before dynamic；

   The memory module is used to store the data to virtual machine shared buffer memory device, when concurrently starting for other virtual machines The data are read from the virtual machine shared buffer memory device；

   Described device also includes deduplication module, and the deduplication module is used for virtual machine shared buffer memory device storage Data carry out data de-duplication operations.
5. 5. device according to claim 4, it is characterised in that described device also includes：Module for reading and writing；

   The module for reading and writing is used to, when the virtual machine sends read-write requests to the virtual machine shared buffer memory device, receive institute Request is stated, and the read-write requests response of the virtual machine shared buffer memory device transmission is sent to the virtual machine, the read-write please Response is asked to carry read-write requests failure information.
6. 6. the device according to claim 4 or 5, it is characterised in that the virtual machine start-up course includes the virtual machine Start shooting to the VME operating system start completion or virtual machine start and arrive the default application program of the virtual machine Start completion.