WO2014100954A1

WO2014100954A1 - Method and system for data controlling

Info

Publication number: WO2014100954A1
Application number: PCT/CN2012/087338
Authority: WO
Inventors: 薄海
Original assignee: 华为技术有限公司
Priority date: 2012-12-24
Filing date: 2012-12-24
Publication date: 2014-07-03
Also published as: CN103229150A; CN103229150B

Abstract

The present invention provides a method and system for data controlling and relates to the technical field of computer. The method includes: receiving a request to read data sent by a cloned virtual machine; determining which mirror the data block requested by the data reading request belongs to; when the data block requested by the reading request belongs to a base mirror, reading the data block from the memory, and sending the data block to the cloned virtual machine; when the data block requested by the reading request belongs to an update mirror, reading the data block from a disk, and sending the data block to the memory, then reading the data block from the memory, and sending the data block to the cloned virtual machine.

Description

Data control method and system

The present invention relates to the field of computer technologies, and in particular, to a data control method and system. Background technique

Currently, virtualization technology can virtualize the physical resources of one physical machine (that is, a physical computer) into multiple virtual machines, and each virtual machine can implement the functions of the physical machine. The virtual machine has a cloning function. One virtual machine (the virtual machine is equivalent to the source virtual machine) can clone one or more cloned virtual machines. Each cloned virtual machine can have the same operating system and application software as the source virtual machine. As well as data files, after the cloned virtual machine is generated, you can also add new data or update existing data.

The prior art provides a link cloning technology for implementing a clone virtual machine operation through a base image and an update image. The base image is the data portion common to the cloned virtual machine and the source virtual machine, and the update image is the data portion of the cloned virtual machine that is different from the source virtual machine.

However, the existing base image and update image are saved on the disk. When each cloned virtual machine runs, each cloned virtual machine writes the data in the base image to be read into memory. The basic images written by the machine in the same memory are the same, resulting in low utilization of memory resources. Summary of the invention

The invention provides a data control method and system, which can improve the utilization rate of memory resources.

In order to achieve the above object, the present invention adopts the following technical solutions:

The first aspect provides a data control method, where the base image is saved in the memory, and the update image is saved in the disk, where the method includes:

Receiving a read data request sent by the clone virtual machine; Determining a mirror image of the data block requested to be read by the read data request;

When the data block requested to be read belongs to the base image, the data block is read from the memory, and the data block is sent to the clone virtual machine;

Reading the data block from the disk when the data block requested to be read belongs to the update image, and writing the data block into the memory, and reading the data from the memory Block, and sent to the cloned virtual machine.

With reference to the first aspect, in a first implementation manner of the first aspect, the basic image is compressed and saved in a virtual device in the memory,

When the data block requested to be read belongs to the base image, reading the data block from the memory, and sending the data block to the clone virtual machine includes: when the request is read When the data block belongs to the base image, the data block is read and decompressed from the virtual device of the memory, and the decompressed data block is sent to the clone virtual machine.

In conjunction with the first aspect or the first implementation of the first aspect, in a second implementation of the first aspect, the method further includes:

Receiving a write data request sent by the clone virtual machine;

Determining whether the data requested by the write data request constitutes a complete data block; if the complete data block is formed, writing to the update image, if not forming a complete data block, from the base image And extracting a data block corresponding to the data, updating the data into a data block corresponding to the data, and writing the updated data block into the update image. With reference to the first aspect, or any one of the foregoing implementation manners of the first aspect, in a third implementation manner of the first aspect, the data block in the memory is saved in each page of the memory,

The method further includes:

Regularly scanning each page in the memory;

The hash operation is performed by using the data in each page to obtain the hash value of each page; If the page with the same hash value is greater than two, one of the pages with the same hash value is retained, and the remaining pages are deleted.

In a second aspect, a data control system is provided. The base image is saved in the memory, and the update image is saved in the disk. The system includes:

a receiving module, configured to receive a read data request sent by the clone virtual machine;

a control module, configured to determine a mirror image to which the data block requested to be read by the read data request belongs;

a reading module, configured to: when the control module determines that the data block requested to be read belongs to a basic image, read the data block from the memory; and send a module, configured to send the data block to the Cloning a virtual machine;

The reading module is further configured to: when the control module determines that the data block requested to be read belongs to an update image, read the data block from the disk; and write a module, where the data is used to The block is written into the memory; the reading module is further configured to read the data block from the memory, and send the data block to the clone virtual machine by the sending module.

With reference to the second aspect, in a first implementation manner of the second aspect, the basic image is compressed and saved in a virtual device in the memory,

The reading module is specifically configured to: when the control module determines that the data block requested to be read belongs to a basic image, read and decompress the data block from the virtual device in the memory;

The sending module is specifically configured to send the decompressed data block to the clone virtual machine.

In conjunction with the second aspect or the first implementation of the second aspect, in a second implementation of the second aspect,

The receiving module is further configured to receive a write data request sent by the cloned virtual machine; the control module is further configured to determine whether the data requested by the write data request constitutes a complete data block;

The writing module is further configured to: if the control module determines that the data requested to be written by the write data request constitutes a complete data block, write the update image; the reading module further uses If the control module determines the write data, please If the data requested to be written does not constitute a complete data block, the data block corresponding to the data is taken out from the base image, and the data is updated into a data block corresponding to the data, the writing module Also used to write the updated data block to the update image.

With the second aspect or the foregoing implementation manner of the second aspect, in a third implementation manner of the second aspect, the data block in the memory is saved in each page of the memory,

The system also includes:

a scanning module, configured to periodically scan each page in the memory;

An operation module, configured to perform hash operation by using data in each page, respectively, to obtain a hash value of each page;

The processing module is configured to reserve one of the pages with the same hash value and delete the remaining pages if the pages with the same hash value are greater than two.

In a third aspect, a data control system is provided. The base image is saved in the memory, and the update image is saved in the disk. The system includes:

a receiver, configured to receive a read data request sent by the clone virtual machine;

a processor, configured to determine a mirror image of the data block requested to be read by the read data request;

The processor is further configured to: when determining that the data block requested to be read belongs to a base image, read the data block from the memory; and send, by the transmitter, the data block to the clone virtual machine;

The processor is further configured to: when the data block requested to be read belongs to an update image, read the data block from the disk, and write the data block into the memory, from the The data block is read in memory and sent by the transmitter to the cloned virtual machine.

With reference to the third aspect, in a first implementation manner of the third aspect, the basic image is compressed and saved in a virtual device in the memory,

The processor is specifically configured to: when determining that the data block requested to be read belongs to the basic image, reading and decompressing the data block from the virtual device of the memory; The transmitter is specifically configured to send the decompressed data block to the clone virtual machine.

In conjunction with the third aspect or the first implementation of the third aspect, in a second implementation of the third aspect,

The receiver is further configured to receive a write data request sent by the clone virtual machine; the processor is further configured to determine whether the data requested by the write data request is a complete data block;

The processor is further configured to write into the update image if it is determined that the data requested to be written by the write data request constitutes a complete data block;

The processor is further configured to: if it is determined that the data requested to be written by the write data request does not constitute a complete data block, extract a data block corresponding to the data from the base image, and update the data to the In the data block corresponding to the data, the updated data block is written into the update image.

With reference to the third aspect, or any one of the foregoing implementation manners of the third aspect, in a third implementation manner of the third aspect, the data block in the memory is saved in a page of the memory,

The processor is further configured to periodically scan each page in the memory; respectively perform hash operation by using data in each page to obtain a hash value of each page; if the page with the same hash value is greater than two, Keep one of the pages with the same hash value and delete the remaining pages.

In the embodiment of the present invention, the basic image shared by each cloned virtual machine is stored in the memory. Therefore, when the data block requested by the cloned virtual machine belongs to the basic image, the data read by the request may be directly read from the memory. Piece. In the embodiment of the present invention, the basic image shared by each cloned virtual machine is stored in the memory, that is, a basic image is saved in the memory, and each cloned virtual machine needs to occupy a certain space in the memory in the prior art. The embodiment of the present invention can improve the utilization of memory resources, as compared with the case where the same basic image is saved in the memory. DRAWINGS In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description It is merely some embodiments of the present invention, and those skilled in the art can obtain other drawings according to the drawings without any creative work.

FIG. 1 is a schematic diagram of a flow of reading and writing data of a cloned virtual machine according to an embodiment of the present invention;

2a is a first flowchart of a data control method according to an embodiment of the present invention; FIG. 2b is a second flowchart of a data control method according to an embodiment of the present invention; ;

Figure 3a is a first structural diagram of a data control system according to an embodiment of the present invention; Figure 3b is a second structural diagram of a data control system according to an embodiment of the present invention; Figure 4 is a data control system according to an embodiment of the present invention; The third structural diagram. detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

Embodiments of the present invention provide a data control method, which is applied to a clone virtual machine when reading and writing data from a base image and an update image. The clone virtual machine is obtained by using a source virtual machine for cloning, and the base image is the same data part of the clone virtual machine and the source virtual machine, and the update image is different data of the clone virtual machine and the source virtual machine. section.

The schematic diagram of the process of reading and writing data of the cloned virtual machine shown in FIG. 1 is saved in the memory, and the updated image is saved in the disk.

Among them, the difference between memory and disk is as follows:

Memory An electronic device made up of semiconductor technology used to store data. The memory capacity is usually small (for example, 16M, 1 G, etc.), but read The speed is faster. When the memory is powered up, the data can be saved, but after the power is turned off, the saved data will be cleared, so it is also called a high-speed temporary storage of information hardware.

A disk can include a hard disk, an electronic device made of magnetic material that stores and retrieves data on a flat magnetic surface. The information is written by the electromagnetic flow to the disk by the write head that is close to the magnetic surface, and the polarity is changed by the electromagnetic flow. The information can be read back in the opposite way. The capacity of a disk is usually large (for example, 40G, 100G, etc.), but the reading speed is slow. A disk is a kind of non-volatile storage device. When the power is turned off, the saved data is not lost. Therefore, it is also called a low-speed hardware that stores information for a long time.

The data saved in the base image includes: an operating system, an application, and a data file cloned from the source virtual machine, where the data saved in the update image includes: a new data added by the clone virtual machine or a base image The data obtained after the data is updated. Each cloned virtual machine has its own update image, but shares the same base image. When each cloned virtual machine starts up, it needs to read data from the base image and read and write data to the respective update image according to their respective requirements.

In the following, a clone virtual machine is taken as an example to describe a data control method. The execution body of the method is a data control system, which can process read and write data requests from the cloned virtual machine.

As shown in FIG. 2a, the method includes:

201. Receive a read data request sent by the clone virtual machine.

202. Determine, by the read data request, a mirror to which the data block requested to be read belongs.

The data block is a minimum unit for performing data read and write operations. Generally, one data block includes 512 bytes.

The data block read by the request may be modified to determine whether the data block read by the request belongs to a basic image or an updated image.

Since the data block in the base image is shared by each clone virtual machine, it is usually not modified. Therefore, when the base image is saved in the memory, the data control system can set the modification flag of each data block in the base image to be unmodified. . The data block saved in the update image is a new data block added by the clone virtual machine or a data block obtained by updating the data block in the base image, and the cloned virtual machine writes the data block to the update image when the data is written. The control system can set the modification flags for these data blocks to have been modified.

It should be noted that, in this paper, the unmodified and modified representation of the modified mark of the data block is a relative concept, mainly to distinguish whether the data block belongs to the basic image or the updated image, that is, the data block marked as unmodified belongs to the basic Mirrored, and the data block marked as modified belongs to the update image.

203. When the data block requested to be read belongs to a base image, read the data block from the memory, and send the data block to the clone virtual machine.

204. When the data block requested to be read belongs to an update image, read the data block from the disk, write the data block into the memory, and read the memory from the memory. The data block is described and sent to the clone virtual machine.

In the embodiment of the present invention, the basic image shared by each cloned virtual machine is stored in the memory. Therefore, when the data block requested by the cloned virtual machine belongs to the basic image, the data read by the request may be directly read from the memory. Piece. In the embodiment of the present invention, the basic image shared by each cloned virtual machine is stored in the memory, that is, a basic image is saved in the memory, and each cloned virtual machine needs to occupy a certain space in the memory in the prior art. The embodiment of the present invention can improve the utilization of memory resources, as compared with the case where the same basic image is saved in the memory.

In addition, in the prior art, since each cloned virtual machine needs to read the data in the base image from the disk, the read/write capability of the disk (the number of input/output (I/O) read/write per unit time) is usually It is fixed, thus limiting the number of cloned virtual machines running simultaneously. The cloning virtual machine in the embodiment of the present invention does not need to read and write the disk when reading the basic image, and only reads and writes the disk when the updated image is read, thereby reducing the number of reading and writing to the disk, and further Support for more cloned virtual machine operations.

In addition, the prior art also needs to manually set the size of the cache required for cloning a virtual machine, which requires high technical and experience of the technician, and adopts an embodiment of the present invention. The method for providing does not need to preset the size of the buffer required for cloning the virtual machine, and can further improve the practical scope of the embodiment of the present invention.

The method provided by the embodiment of the present invention can also process the write data request from the clone virtual machine. details as follows:

Receiving a write data request sent by the clone virtual machine; determining whether the data requested by the write data request constitutes a complete data block. If the complete data block is formed, it is written into the update image; if the complete data block is not formed, the data block corresponding to the data is taken out from the base image, and the data is updated to the data corresponding In the data block, the updated data block is written into the update image.

Since the data block is the smallest unit for reading and writing data operations, the data written by the cloned virtual machine does not necessarily constitute a complete data block. For example, the data that the cloned virtual machine needs to write is in a certain data block. Part of the byte is updated. Therefore, when the part of the byte is written to the update image, the data block corresponding to the partial byte needs to be read from the base image, and the partial byte is updated to the At the location of the corresponding byte in the data block, for example, the clone virtual machine needs to write 20 bytes and use the 20 bytes to update the 100-120 bytes of a data block, then from the base image After reading the data block, updating the 100-120 bytes of the data block to the 20 bytes, and writing the updated data block into the update image, specifically requiring writing The cloned virtual machine is in the update image.

Further, in the method provided by the embodiment of the present invention, the basic image may be compressed and saved in the virtual device of the memory, so as to save memory resources.

In the step 203, when the data block requested to be read belongs to the basic image, the data block is read from the memory, and the data block is sent to the clone virtual machine, and the following steps are specifically replaced. Execution: When the data block read by the request belongs to the basic image, the data block is read and decompressed from the virtual device of the memory, and the decompressed data block is sent to the clone virtual machine.

Among them, you can use the ZRAM tool to compress the base image and save it in the virtual device of the memory. The ZRAM tool can create a virtual device in memory. The format of the virtual device can be ext4, ext2 or XFS. Save the base image in In the virtual device, the ZRAM tool will automatically compress the base image by using a compression algorithm and save it in the virtual device. The compression algorithm employed is for example the LZO algorithm. When the base image is read from the virtual device, the ZRAM tool will automatically decompress the base image using the same compression algorithm and read the decompressed data.

It should be noted that the base image is stored in a virtual device in the memory, and all the saved data is cleared when the memory is powered off. Therefore, when the memory is powered on again, the ZRAM tool needs to be used again to perform the base image. Saved in a virtual device in memory.

Understandably, after the ZRAM tool is used to save the base image in a virtual device in the memory, if the memory is not powered off, the startup and shutdown of each virtual machine will not affect the base image saved in the memory. When the memory is powered off, you need to use the ZRAM tool to save the base image in a virtual device in memory when the power is turned back on.

Further, the data block in the memory is stored in each page of the memory, that is, in the data block in the base image stored in advance in the memory, or in the update image read from the disk. When a data block is written to the internal memory, the memory is used to save the page in memory.

The memory management unit divides the virtual address space of the memory in the form of a page. The size of the page in memory can be 2 to the power of n.

Since the data in the update image of the cloned virtual machine and the data in the base image may be the same, for example, an application in the base image of the cloned virtual machine and an application in the update image use the same link library. Therefore, the page in which the link library is stored in the memory will be duplicated; or, since the data in the updated image of each cloned virtual machine may also have the same situation, the image will be updated in each of the cloned virtual machines. When data is written to memory, it also causes problems with multiple pages holding the same data.

To further improve the utilization of memory resources, and effectively save memory resources, as shown in FIG. 2b, the method further includes the following steps:

205. Regularly scan each page in the memory; 206, respectively, using the data in each page to perform a hash operation, and obtaining a hash value of each page;

207. If the pages with the same hash value are greater than two, one page of the page with the same hash value is reserved, and the remaining pages are deleted.

It should be noted that, the data control system records the calling end of each page, and the calling end is an application that needs to invoke data in the page, etc., when the data control system receives the request sent by the virtual machine to delete data in a certain page. For example, when an application that uses a certain page in a cloned virtual machine exits, the page needs to be deleted from the memory, and a request to delete data in a certain page is sent to the data control system, and the data control system determines Whether there are other callers on the page. If it exists, the operation of deleting the page is not performed. Otherwise, the page is deleted from the memory.

As shown in Figure 2c, the page share diagram, the base image, the update image of the clone VM VM1, and the update mirror of the clone VM VM2 have the same data block (the slashes represent the same data block), and are only retained in memory. For each page in which the same data block is stored, the update image of the clone virtual machine VM1, the update image of the clone virtual machine VM2, and the base image all point to the one page through the page pointer. When the cloned virtual machine needs to update the data block in the image by itself, the page pointer can be used to determine the page where the required data block is stored. Through page sharing, the embodiment of the present invention can further improve the utilization of memory resources.

The embodiment of the present invention further provides a data control system. The base image is saved in the memory, and the update image is saved in the disk. For the basic image and the update image, refer to the description of the virtual machine read/write data flow shown in FIG.

As shown in Figure 3a, the system includes:

The receiving module 31 is configured to receive a read data request sent by the clone virtual machine, where the control module 32 is configured to determine a mirror image of the data block requested by the read data request, where the data block is for reading and writing data. The smallest unit, usually, a block of data consists of 512 bytes.

The control module may determine whether the data block read by the request belongs to a base image or an update image according to whether the data block read by the request is modified. Specific can be See the related description in step 202 of the above method embodiment.

a reading module 33, configured to: when the control module determines that the data block requested to be read belongs to the basic image, read the data block from the memory; and send a module 34, configured to: a block is sent to the clone virtual machine;

The reading module 33 is further configured to: when the control module determines that the data block requested to be read belongs to the update image, read the data block from the disk; and write a module 35, configured to: Writing the data block into the memory; the reading module 33 is further configured to read the data block from the memory, and send the data block to the clone virtual machine by the sending module.

In addition, in the prior art, since each cloned virtual machine needs to read the data in the base image from the disk, the read/write capability of the disk (the number of input/output (I/O) read/write per unit time) is usually It is fixed, thus limiting the number of cloned virtual machines running simultaneously. In the system provided by the embodiment of the present invention, the cloned virtual machine does not need to read and write the disk when reading the base image, and only reads and writes the disk when the update image is read, thereby reducing read and write of the disk. The number of times, which in turn supports the operation of more cloned virtual machines.

In addition, the prior art also needs to manually set the size of the cache required for cloning the virtual machine. The technical and experience of the technician is high. The system provided by the embodiment of the present invention does not need to preset the cache required for the cloned virtual machine. The size can further improve the practical range of the embodiments of the present invention. Further, when the system can also request write data from the cloned virtual machine Processing, specifically:

The receiving module 31 is further configured to receive a write data request sent by the cloned virtual machine.

The control module 32 is further configured to determine whether the data requested by the write data request forms a complete data block;

The writing module 35 is further configured to: if the control module determines that the data requested to be written by the write data request constitutes a complete data block, write the update image; the reading module 33, And if the control module determines that the data requested to be written by the write data request does not constitute a complete data block, extracting a data block corresponding to the data from the base image, and updating the data to In the data block corresponding to the data, the writing module 35 is further configured to write the updated data block into the update image.

Since the data block is the smallest unit for reading and writing data operations, the data written by the cloned virtual machine does not necessarily constitute a complete data block. For example, the data that the cloned virtual machine needs to write is in a certain data block. Part of the byte is updated. Therefore, when the part of the byte is written to the update image, the data block corresponding to the partial byte needs to be read from the base image, and the partial byte is updated to the At the location of the corresponding byte in the data block, for example, the clone virtual machine needs to write 20 bytes and use the 20 bytes to update the 100-120 bytes of a data block, then the data is After the block is read, the 100-120 bytes of the data block are updated to the 20 bytes, and the updated data block is written into the update image, specifically written into the clone virtual The machine is updated in the image.

Further, the base image is compressed and stored in the memory, wherein the base image can be compressed and saved in a virtual device in the memory by using a ZRAM tool.

The reading module 33 is specifically configured to: when the control module determines that the data block requested to be read belongs to the basic image, read and decompress the data block from the virtual device in the memory;

The sending module 34 is specifically configured to send the decompressed data block to the clone virtual machine. Further, the data blocks in the memory are saved in each page of the memory, so as to further improve the utilization of memory resources, effectively saving memory resources, as shown in the figure.

As shown in 3b, the system further includes:

The scanning module 36 is configured to periodically scan each page in the memory;

The operation module 37 is configured to perform hash operation by using data in each page to obtain a hash value of each page;

The processing module 38 is configured to reserve one page of the page with the same hash value if the number of pages with the same hash value is greater than two, and delete the remaining pages.

It should be noted that the data control system may also record the calling end of each page, where the calling end is an application that needs to invoke data in the page, etc., when the data control system receives the data sent by the virtual machine and deletes a certain page. When requesting, for example, when an application that uses a page in a cloned virtual machine exits, the page needs to be deleted from the memory, and a request to delete data in a page is sent to the data control system, the data control The system determines whether there are other callers on the page. If it exists, the operation of deleting the page is not performed. Otherwise, the page is deleted from the memory.

The system shown in Figure 3b can also be referred to the relevant description in the method described in Figures 2b and 2c.

The embodiment of the present invention further provides a data control system, wherein the base image is saved in the memory, and the update image is saved in the disk. For the basic image and the update image, refer to the description of the virtual machine read/write data flow shown in FIG. .

As shown in FIG. 4, the data control system includes: a processor 41, a receiver 42, a transmitter 43, a memory 44, a bus 4000, and a driving circuit 4001.

The receiver 42 is configured to receive a read data request sent by the clone virtual machine, and the processor 41 is configured to determine a mirror image to which the read data request is requested to be read;

The processor 41 is further configured to: when determining that the data block requested to be read belongs to the basic image, read the data block from the memory; and the transmitter 43 is configured to send the data block to the Cloning a virtual machine;

The processor 41 is further configured to: when the data block requested to be read belongs to an update image Reading the data block from the disk and writing the data block into the memory, reading the data block from the memory, and transmitting by the transmitter 43 to the Clone a virtual machine.

The driving circuit 4001 is configured to provide drivers for each hardware in the data control system so that each hardware can work normally.

In a specific implementation of the present embodiment, the memory includes at least one or more of the following memory devices, a read only memory, a random access memory or a nonvolatile random access memory, the memory providing instructions and data to the processor.

Wherein, the processor may be an integrated circuit chip with signal processing capability. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in the processor or an instruction in a form of software. These instructions may be implemented and controlled by a processor therein for performing the methods disclosed in the embodiments of the present invention. The processor may also be a general purpose processor, a digital signal processing (DSP), an application specific integrated circuit, a Field Programmable Gate Array (FPGA) or other programmable logic. Devices, discrete gates or transistor logic devices, discrete hardware components.

The above general purpose processor may be a microprocessor or the processor may be any conventional processor, decoder or the like. The steps of the method disclosed in the embodiment of the present invention may be directly implemented as a hardware processor, or may be performed by a combination of hardware and software modules in the processor. The software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.

Additionally, the various hardware components of the data control system are coupled together by a bus system 4000, which in addition to the data bus includes a power bus, a control bus, and a status signal bus. However, for clarity of description, various buses are labeled as bus system 4000 in FIG.

In the embodiment of the present invention, the basic image shared by each cloned virtual machine is stored in the memory. Therefore, when the data block requested by the cloned virtual machine belongs to the basic image, Reading the data block read by the request directly from memory. In the embodiment of the present invention, the basic image shared by each cloned virtual machine is stored in the memory, that is, a basic image is saved in the memory, and each cloned virtual machine needs to occupy a certain space in the memory in the prior art. The embodiment of the present invention can improve the utilization of memory resources, as compared with the case where the same basic image is saved in the memory.

In addition, in the prior art, since each cloned virtual machine needs to read the data in the base image from the disk, the read/write capability of the disk (the number of input/output (I/O) read/write per unit time) is usually It is fixed, thus limiting the number of cloned virtual machines running simultaneously. In the system provided by the present invention, the cloned virtual machine does not need to read and write the disk when reading the base image, and only reads and writes the disk when the update image is read, thereby reducing the number of reads and writes to the disk. In turn, it supports the operation of more cloned virtual machines.

In addition, the prior art also needs to manually set the size of the cache required for cloning the virtual machine. The technical and experience of the technician is high. The system provided by the embodiment of the present invention does not need to preset the cache required for the cloned virtual machine. The size can further improve the practical range of the embodiments of the present invention. The system provided by the embodiment of the present invention can also process the write data request from the cloned virtual machine. details as follows:

The receiver 42 is further configured to receive a write data request sent by the clone virtual machine, where the processor 41 is further configured to determine whether the data requested by the write data request constitutes a complete data block;

The processor 41 is further configured to write into the update image if it is determined that the data requested to be written by the write data request constitutes a complete data block;

The processor 41 is further configured to: if it is determined that the data requested to be written by the write data request does not constitute a complete data block, extract a data block corresponding to the data from the base image, and update the data to the In the data block, the updated data block is written into the update image. Further, the base image is compressed and stored in the memory 44. Among them, you can use the ZRAM tool to compress the base image and save it in the virtual device of the memory.

The processor 41 is specifically configured to: when determining that the data block requested to be read belongs to the basic image, reading and decompressing the data block from the virtual device in the memory; The decompressed data block is sent to the clone virtual machine.

Further, the data block in the memory is saved in the page of the memory, in order to further improve the utilization of the memory resource, and effectively save the memory resource, the processor 41 is further configured to periodically scan each of the memory. a page; respectively, using the data in each page to perform a hash operation to obtain a hash value of each page; if the number of pages having the same hash value is greater than two, one page of the same hash value is retained, and the rest is deleted. Page. Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general hardware, and of course, by hardware, but in many cases, the former is a better implementation. . Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a readable storage medium, such as a floppy disk of a computer. A hard disk or optical disk or the like includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments of the present invention.

The above is only the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the present invention. It should be covered by the scope of the present invention. Therefore, the scope of the invention should be determined by the scope of the appended claims.

Claims

claims

1. A data control method, characterized in that the basic image is stored in the memory, and the updated image is stored in the disk. The method includes:

Receive data read requests sent by cloned virtual machines;

Determine the image to which the data block requested to be read by the read data request belongs;

When the data block requested to be read belongs to the base image, read the data block from the memory and send the data block to the clone virtual machine;

When the data block requested to be read belongs to the update image, the data block is read from the disk, written into the memory, and then read from the memory. The data block is then sent to the cloned virtual machine.

2. The method according to claim 1, characterized in that the basic image is compressed and stored in the virtual device of the memory,

When the data block requested to be read belongs to the base image, reading the data block from the memory and sending the data block to the clone virtual machine includes: When the fetched data block belongs to the base image, the data block is read from the virtual device in the memory and decompressed, and the decompressed data block is sent to the clone virtual machine.

3. The method according to claim 1 or 2, further comprising: receiving a data write request sent by the cloned virtual machine;

Determine whether the data requested by the write data request constitutes a complete data block; if it constitutes a complete data block, write it into the update image; if it does not constitute a complete data block, then retrieve it from the base image The data block corresponding to the data is taken out, the data is updated into the data block corresponding to the data, and the updated data block is written into the update mirror.

4. The method according to any one of claims 1 to 3, characterized in that the data blocks in the memory are stored in each page of the memory,

The method also includes:

Periodically scan various pages in said memory;

Use the data in each page to perform hash operations to obtain the hash of each page. value;

If there are more than two pages with the same hash value, keep one of the pages with the same hash value and delete the remaining pages.

5. A data control system, characterized in that the basic image is stored in the memory and the update image is stored in the disk. The system includes:

The receiving module is used to receive the read data request sent by the cloned virtual machine;

A control module, used to determine the mirror to which the data block requested to be read by the read data request belongs;

A reading module, configured to read the data block from the memory when the control module determines that the data block requested to be read belongs to the basic image; a sending module, used to send the data block Give the cloned virtual machine;

The reading module is also configured to read the data block from the disk when the control module determines that the data block requested to be read belongs to the update image; the writing module is used to write the data block to the disk. The data block is written into the memory; the reading module is also configured to read the data block from the memory, and send the data block to the clone virtual machine by the sending module.

6. The system according to claim 5, wherein the basic image is compressed and stored in the virtual device of the memory,

The reading module is specifically configured to read and decompress the data block from the virtual device of the memory when the control module determines that the data block requested to be read belongs to the basic image;

7. The system according to claim 5 or 6, characterized in that,

The receiving module is also used to receive the write data request sent by the clone virtual machine; the control module is also used to determine whether the data requested by the write data request constitutes a complete data block;

The writing module is also configured to write into the update image if the control module determines that the data requested by the write data request constitutes a complete data block;

The reading module is also used to determine if the control module determines that the write data request If the data requested to be written does not constitute a complete data block, the data block corresponding to the data is taken out from the basic image, and the data is updated to the data block corresponding to the data. The writing module, It is also used to write updated data blocks into the updated image.

8. The system according to any one of claims 5-7, characterized in that the data blocks in the memory are stored in each page of the memory,

The system also includes:

A scanning module for regularly scanning each page in the memory;

The operation module is used to perform hash operations using the data in each page to obtain the hash value of each page;

A processing module configured to retain one of the pages with the same hash value and delete the remaining pages if there are more than two pages with the same hash value.

9. A data control system, characterized in that the basic image is stored in the memory and the updated image is stored in the disk. The system includes:

Receiver, used to receive read data requests sent by the cloned virtual machine;

The processor is configured to determine the mirror to which the data block requested to be read by the read data request belongs; the processor is also configured to read from the memory when it is determined that the data block requested to be read belongs to the basic mirror. Get the data block; Transmitter, used to send the data block to the clone virtual machine;

The processor is also configured to read the data block from the disk and write the data block into the memory when the data block requested to be read belongs to the update image. The data block is read from the memory and sent to the clone virtual machine by the transmitter.

10. The system according to claim 9, characterized in that the basic image is compressed and stored in the virtual device of the memory,

The processor is specifically configured to read and decompress the data block from the virtual device of the memory when it is determined that the data block requested to be read belongs to the basic image;

The transmitter is specifically used to send decompressed data blocks to the clone virtual machine.

11. The system according to claim 9 or 10, characterized in that,

The receiver is also used to receive a write data request sent by the clone virtual machine; The processor is also used to determine whether the data requested to be written by the write data request constitutes a complete data block;

The processor is also configured to write into the update image if it is determined that the data requested by the write data request constitutes a complete data block;

The processor is also configured to, if it is determined that the data requested by the write data request does not constitute a complete data block, retrieve the data block corresponding to the data from the base image, and update the data. to the data block corresponding to the data, and write the updated data block into the update mirror.

12. The system according to any one of claims 9-11, characterized in that the data blocks in the memory are stored in each page of the memory,

The processor is also used to regularly scan each page in the memory; perform a hash operation using the data in each page to obtain the hash value of each page; if there are more than two pages with the same hash value, then Keep one of the pages with the same hash value and delete the rest.