WO2023160397A1

WO2023160397A1 - Memory management methods and apparatuses

Info

Publication number: WO2023160397A1
Application number: PCT/CN2023/075100
Authority: WO
Inventors: 赵三谦; 王阳杰
Original assignee: 阿里巴巴（中国）有限公司
Priority date: 2022-02-25
Filing date: 2023-02-09
Publication date: 2023-08-31
Also published as: CN114579304A

Abstract

Provided in the embodiments of the present description are memory management methods and apparatuses. One memory management method comprises: receiving a memory release request in respect of a target virtual machine; in response to the memory release request, determining at least one memory block to be released corresponding to the target virtual machine, and determining memory block attribute information corresponding to each memory block to be released; determining a memory block to be released, the attribute information of which is of a first type, as a first-type memory block to be released; and releasing the memory of the first-type memory block to be released by adjusting refresh configuration parameters of the first-type memory block to be released. By means of the method, release and clearing of the memory can be rapidly completed by controlling the refreshing frequency of the first-type memory block to be released. In addition, the mode of adjusting the refreshing frequency of the memory block to be released occupies few resources and has little influence on other running virtual machines.

Description

Memory management method and device

This application claims the priority of the Chinese patent application with the application number 202210180906.0 and the application name "memory management method and device" submitted to the China Patent Office on February 25, 2022, the entire contents of which are incorporated in this application by reference.

technical field

The embodiments of this specification relate to the technical field of computers, and in particular to memory management methods.

Background technique

In a cloud computing scenario based on virtualization technology, in order to ensure data security when the virtual machine is shut down, the memory used by it needs to be released and cleared. and memory bandwidth resources. With the rapid development of cloud computing technology, the memory specifications of virtual machines are getting larger and larger. When a virtual machine is shut down, the corresponding memory release and clearing delay and high resource usage will be It has a greater impact on other virtual machines in use.

Contents of the invention

In view of this, the embodiment of this specification provides a memory management method. One or more embodiments of this specification also relate to a memory management device, a computing device, a computer-readable storage medium and a computer program, so as to solve technical defects existing in the prior art.

According to the first aspect of the embodiments of this specification, a memory management method is provided, including:

Receive a memory release request for the target virtual machine;

Determining at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determining memory block attribute information corresponding to each memory block to be released;

Determining that the memory block attribute information is the first type of memory block to be released is the first type of memory block to be released, wherein the first type of memory block to be released is a complete memory block;

The memory of the first type of memory block to be released is released by adjusting the refresh configuration parameters of the first type of memory block to be released.

According to the second aspect of the embodiments of this specification, a memory management method is provided, including:

receiving a memory allocation request for the target virtual machine, wherein the memory allocation request carries memory information to be allocated;

Obtain the number of free memory blocks and attribute information of free memory blocks;

Allocating a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block, wherein the first type of free memory block is a complete memory block .

According to a third aspect of the embodiments of this specification, a memory management device is provided, including:

a receiving module configured to receive a memory release request for the target virtual machine;

A response module, configured to determine at least one waiting memory corresponding to the target virtual machine in response to the memory release request Release the memory block, and determine the memory block attribute information corresponding to each memory block to be released;

The determining module is configured to determine that the attribute information of the memory block is that the first type of memory block to be released is the first type of memory block to be released, wherein the first type of memory block to be released is a complete memory block;

The release module is configured to release the memory of the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released.

According to a fourth aspect of the embodiments of this specification, a memory management device is provided, including:

An allocation request receiving module configured to receive a memory allocation request for the target virtual machine, wherein the memory allocation request carries memory information to be allocated;

An information acquisition module configured to acquire the number of free memory blocks and attribute information of the free memory blocks;

An allocation module configured to allocate a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block, wherein the first type of free memory block A memory block is a complete memory block.

According to a fifth aspect of the embodiments of this specification, a computing device is provided, including:

memory and processor;

The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the above memory management method are implemented.

According to a sixth aspect of the embodiments of the present specification, there is provided a computer-readable storage medium, which stores computer-executable instructions, and implements the steps of the memory management method above when the instructions are executed by a processor.

According to a seventh aspect of the embodiments of the present specification, a computer program is provided, wherein, when the computer program is executed in a computer, it causes the computer to execute the steps of the memory management method above.

An embodiment of this specification realizes receiving a memory release request for a target virtual machine; determining at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determining the memory corresponding to each memory block to be released Block attribute information; determine that the memory block attribute information is the first type of memory block to be released as the first type of memory block to be released; release the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released The memory of the memory block. Through this method, by controlling the refresh frequency of the first type of memory blocks to be released, the release and clearing of the memory can be quickly completed, and at the same time, by adjusting the refresh frequency of the memory blocks to be released, less resources are occupied, and other running virtual machine is less affected.

Description of drawings

FIG. 1 is a flowchart of a memory management method provided by an embodiment of this specification;

Fig. 2 is a flow chart of a memory management method provided by another embodiment of this specification;

FIG. 3 is a flow chart of the processing process of a memory management method applied to a virtual machine M to allocate memory provided by an embodiment of the present specification;

FIG. 4 is a flow chart of a processing process of a memory management method applied to a virtual machine M to release memory provided by an embodiment of this specification;

Fig. 5 is a schematic structural diagram of a memory management device provided by an embodiment of this specification;

Fig. 6 is a schematic structural diagram of a memory management device provided by another embodiment of this specification;

Fig. 7 is a structural block diagram of a computing device provided by an embodiment of this specification.

Detailed ways

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the specification. However, this specification can be implemented in many other ways different from those described here, and those skilled in the art can make similar extensions without violating the connotation of this specification, so this specification is not limited by the specific implementations disclosed below.

Terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only, and are not intended to limit one or more embodiments of this specification. As used in one or more embodiments of this specification and the appended claims, the singular forms "a", "the", and "the" are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and/or" used in one or more embodiments of the present specification refers to and includes any or all possible combinations of one or more associated listed items.

It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present specification, the information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, the first may also be referred to as the second, and similarly, the second may also be referred to as the first without departing from the scope of one or more embodiments of the present specification. Depending on the context, the word "if" as used herein may be interpreted as "at" or "when" or "in response to a determination."

First, terms and terms involved in one or more embodiments of this specification are explained.

RAM: (Random access memory) dynamic random access memory, also known as memory.

Bank: (the storage sub-block in the memory chip, each sub-block has multiple word lines, and each word line is connected to multiple data storage units. For example, if there are 8 Banks in an 8G memory, the size of each Bank is 1G, the Bank on one memory shares the IO of one memory. The reading and writing in each Bank can be processed in parallel.

Refresh: The memory needs to be refreshed periodically (within a few milliseconds), otherwise the data stored in the memory will be lost.

Partial Array Self Refresh: (Partial Array Self Refresh, PASR), a memory controller technology that can configure the self-refresh parameters of each memory sub-block, and can refresh the memory sub-blocks completely independently.

PASR memory controller driver: It runs in the kernel of the operating system and provides access to the control registers in the memory controller to dynamically adjust the refresh settings of the memory block.

In a cloud computing scenario based on virtualization technology, in order to ensure data security when a virtual machine is shut down, the memory used by it needs to be cleared by the Hypervisor. Memory clearing requires a large amount of CPU computing power and memory bandwidth resources of the host machine. With the development of cloud computing technology, the memory specifications of virtual machines are getting larger and larger, and a single reset requires longer delay and higher resource usage. Therefore, the memory of large-scale virtual machines is cleared when it is shut down. The impact of the operation on other virtual machines in use is also increasing, such as the failure of new virtual machine scheduling due to memory not being released, and the clearing task of the host machine competing for resources with other virtual machines. Therefore, how to efficiently It is more and more important to clear memory with low consumption.

In this specification, a memory management method is provided. This specification also relates to a memory management device, a computing device, and a computer-readable storage medium, which will be described in detail in the following embodiments.

Referring to FIG. 1 , FIG. 1 shows a flow chart of a memory management method provided according to an embodiment of the present specification, which specifically includes the following steps.

Step 102: Receive a memory release request for the target virtual machine.

Wherein, the target virtual machine specifically refers to a virtual machine that needs to perform memory release and clearing, and the virtual machine runs in the host machine and occupies resources such as CPU and memory of the host machine. The memory release request specifically refers to a request for releasing and clearing the memory occupied by the target virtual machine, releasing the memory occupied by the virtual machine, and specifically refers to clearing the memory occupied by the virtual machine.

In practical applications, when shutting down a virtual machine, the memory of the host machine occupied by the virtual machine needs to be cleared, then the virtual machine is the target virtual machine, and when a shutdown request is received, a memory release request is triggered at the same time.

In a specific implementation provided in this specification, take the host machine with 16G of memory as an example, and there are 4 virtual machines running: virtual machine 1, virtual machine 2, virtual machine 3, and virtual machine 4. At this time, close the virtual machine machine 3, the virtual machine 3 is the target virtual machine, and triggers a memory release request for the memory of the host machine occupied by the virtual machine 3.

Step 104: Determine at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determine memory block attribute information corresponding to each memory block to be released.

Among them, the memory block is the Bank, and the Bank is the storage sub-block on the memory. If a memory is 8G and there are 8 Banks, then a Bank is 1G.

In practical applications, when a virtual machine runs in the host machine, it will occupy the memory and CPU resources of the host machine. For example, if a virtual machine needs 4G memory to run normally, it needs to allocate 4G memory resources from the host machine for the virtual machine. Use, further, the 4G memory resources are distributed in the Bank on the host memory, when the virtual machine is shut down, the 4G memory resources occupied by the virtual machine need to be released, that is, the The Bank corresponding to the 4G memory is released. The Bank corresponding to the 4G memory is the memory block to be released. Since the memory specification of the virtual machine is getting larger and larger, it occupies more and more banks of the host memory. Specifically, a virtual machine The number of memory blocks corresponding to the actual application shall prevail.

In a specific implementation provided in this manual, following the above example, the memory of the host machine is 16G, and there are 16 memory blocks (Banks) in total, and the memory capacity of each Bank is 1G, and the virtual machine 3 occupies 4G of memory. According to The memory release request determines the five banks corresponding to the virtual machine 3, namely Bank3, Bank4, Bank5, Bank6, and Bank7.

In practical applications, each memory block corresponds to an address decoder and a column address decoder, and each memory block corresponds to a corresponding physical address. In simple terms, for an 8G memory, there are 8 memory blocks in total, and memory block 1 corresponds to The physical address of memory block 2 is 1-1G, the physical address corresponding to memory block 2 is 1G-2G, ... and so on. Therefore, the corresponding memory block can be determined according to the corresponding physical address. Based on this, determining at least one memory block to be released corresponding to the target virtual machine in response to the memory release request includes:

determining a virtual machine physical address range corresponding to the target virtual machine in response to the memory release request;

Determine the memory block corresponding to the physical address range of the virtual machine as the memory block to be released.

After receiving the memory release request, the virtual machine physical address range corresponding to the target virtual machine can be determined. The virtual machine physical address range specifically refers to the physical address range occupied by the target virtual machine when it is running. When the virtual machine starts, it will According to the memory usage of the host machine, the virtual machine physical address range is allocated to the virtual machine. Correspondingly, when the memory release request is received, the virtual machine physical address range corresponding to the target virtual machine can be determined. According to the virtual machine physical address range, that is It can be determined that the corresponding memory block is a memory block to be released.

It should be noted that in actual applications, due to different application environments, there may be multiple virtual machine physical address intervals corresponding to the target virtual machine, that is, there are multiple virtual machine physical address intervals corresponding to a target virtual machine, for example, a certain The virtual machine physical address interval corresponding to the virtual machine is 0-140M, 895M-1348M, 1972M-3451M...etc., therefore, each virtual machine physical address interval may correspond to one or more memory blocks. Therefore, at least one memory block to be released can be determined according to the physical address range of the virtual machine.

After at least one memory block to be released is determined, the memory block attribute information of each memory block to be released can be determined. In the embodiment provided in this specification, the attribute information of the memory block is used to express whether the memory block to be released belongs to Exclusive to the target virtual machine. For example, if only the target virtual machine is running on a certain memory block, the attribute information of the memory block is the first type. For other virtual machines, the memory block attribute information of the memory block is the second type.

Specifically, determine the memory block attribute information corresponding to each memory block to be released, including:

determining a target memory block to be released in the at least one memory block to be released;

Determine the physical address range of the memory block of the target memory block to be released;

Judging whether there is a memory block physical address in the physical address range of the memory block that does not belong to the physical address range of the virtual machine;

If not, then determine that the memory block attribute information of the target memory block to be released is the first type;

If yes, determine that the memory block attribute information of the target memory block to be released is the second type.

In practical applications, there is at least one memory block to be released, and each memory block to be released needs to be judged to determine the memory block attribute information of the memory block to be released. At this time, the memory block to be released whose attribute information is being determined is block is the target memory block to be freed.

After the target memory block to be released is determined, the memory block physical address range of the target memory block to be released can be determined, and the memory block physical address range specifically refers to the start physical address and the end physical address of the target memory block to be released The physical address range of the memory block is continuous.

After obtaining the memory block physical interval of the target memory block to be released, it is judged whether there is a memory block physical address that does not belong to the physical address interval of the virtual machine in the memory block physical address interval. If only the target virtual machine runs in the block, the memory block attribute information of the target memory block to be released is the first type; if so, it means that not only the target virtual machine but also other virtual machines are running in the target memory block to be released. machine, it may be determined that the memory block attribute information of the target memory block to be released is the second type.

In a specific implementation method provided in this manual, following the above example, virtual machine 3 corresponds to five Banks, namely Bank3, Bank4, Bank5, Bank6, Bank7, and the virtual machine physical address range of virtual machine 3 is "add3700-add4000, add4000-add5000, add5000-add6000, add6000-add7000, add7200-add7800", the physical address range of the memory block of Bank3 is add3000-add4000, the physical address range of the memory block of Bank4 is add4000-add5000, and the physical address range of the memory block of Bank5 is add5000 -add6000, the physical address range of the memory block of Bank6 is add6000-add7000, and the physical address range of the memory block of Bank7 is add7000-add8000.

Bank3 is the target memory block to be released. In Bank3, the memory block physical address of add3000-add3700 does not belong to the physical address range of the virtual machine. Therefore, the memory block attribute information of Bank3 is the second type; Bank4 is the target to be released For the memory block, add4000-add5000 in Bank4 is the virtual machine physical address range corresponding to virtual machine 3, therefore, the memory block attribute information of Bank4 is the first type..., and so on, the memory block attribute information of Bank3 is the second type , the memory block attribute information of Bank4 is the first type, the memory block attribute information of Bank5 is the first type, the memory block attribute information of Bank6 is the first type, and the memory block attribute information of Bank7 is the second type.

Step 106: Determine that the memory block whose attribute information is the first type to be released is the first type of memory block to be released.

After the attribute information of the memory block is determined, each memory block to be released can be classified according to the attribute information of the memory block. Correspondingly, the memory block to be released whose attribute information of the memory block is the first type is the first type of memory block to be released ; The attribute information of the memory block is that the memory block to be released of the second type is the memory block to be released of the second type. Different memory release policies may be implemented for different types of memory blocks to be released. Specifically, the first type of memory block to be released is a complete memory block, and the second type of memory block to be released is a shared memory block.

In a specific implementation manner provided in this specification, following the above example, Bank4, Bank5, and Bank6 are the first type of memory blocks to be released, and Bank3 and Bank7 are the second type of memory blocks to be released.

Step 108: Release the memory of the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released.

For the first type of memory block to be released, the memory of the first type of memory block to be released can be released and cleared through the first memory release policy, that is, by adjusting the refresh configuration parameters of the first type of memory block to be released.

PASR (Partial Array Self Refresh) is a memory control technology that can configure the self-refresh time of each memory block and can completely independently control each memory block. The memory block needs to be refreshed periodically (within a few milliseconds). If the cycle is not refreshed, the data of the memory block will be cleared and lost.

Based on this principle, after the first type of memory block to be released is determined, the memory of the first type of memory block to be released can be released and cleared by adjusting the refresh configuration parameters of the first type of memory block to be released.

Specifically, release the memory of the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released, including:

Setting the refresh configuration parameter of the first type of memory block to be released to disabled, and counting the duration of the disabled;

When the disabled duration is greater than a preset threshold, the refresh configuration parameter of the first type of memory block to be released is set to enable, wherein the preset threshold is the data storage duration of the memory block to save data.

In practical applications, through the PASR technology, the refresh configuration parameters of the first type of memory block to be released are set to prohibit, that is, no refresh, and the disabled time is counted. When the disabled time exceeds the effective retention time of the data in the memory (that is, the above-mentioned Periodically), then the refresh configuration parameter of the memory block is set to enable. So far, the memory of the first type of memory block to be released has completed the process of releasing and clearing.

In a specific implementation method provided in this specification, following the above example, Bank4, Bank5, and Bank6 are the first type of memory blocks to be released, and the data storage time of each Bank is 50 milliseconds, and the memory controller is called to drive Bank4 through PASR , Bank5, Bank6 refresh configuration parameters are set to disabled, adjust it to no refresh, and count the disabled time length, when the disabled time exceeds 50 milliseconds, then set the Bank4, Bank5, Bank6 refresh configuration parameters to enable, adjust It can be refreshed during this period, and the memory of Bank4, Bank5, and Bank6 has completed the process of releasing and clearing.

In practical applications, the memory block attribute information of some memory blocks is of the second type, and the processing of the second type of memory block attribute information includes:

Determining that the memory block attribute information is the second type of memory block to be released is the second type of memory block to be released;

Obtain the to-be-released physical address range corresponding to each second type of memory block to be released;

Clear the memory of each physical address range to be released.

Corresponding to the memory block whose attribute information of the memory block is the first type, the to-be-released memory block whose attribute information of the memory block is the second type is the second-type to-be-released memory block, and then obtains the corresponding The to-be-released physical address range, the to-be-released physical address range specifically refers to the physical address range occupied by the target virtual machine in the second type of memory block to be released.

After obtaining the physical address range to be released, the memory in the range can be cleared. Specifically, run the code through the Hypervisor, and use the CPU to write 0 to each physical address to be released to clear the physical address to be released. .

The above is the memory release request for the target virtual machine when it is shut down. In practical applications, in order to apply the above-mentioned embodiments provided by this specification, it is preferable to allocate the same memory block for the target virtual machine. Specifically, after receiving the target virtual machine Before the memory release request of the machine, the method further includes S1002-S1006:

S1002. Receive a memory allocation request for the target virtual machine, where the memory allocation request carries memory information to be allocated.

The memory allocation request specifically refers to the request to allocate memory for the target virtual machine. In practical applications, when starting a virtual machine in the host machine, it is necessary to allocate memory for the virtual machine for the virtual machine to start. Based on this, the started virtual machine The machine is the target virtual machine, and the memory allocation request is a request to allocate memory for the target virtual machine. In practical applications, the memory allocation request will carry information about the memory to be allocated, and the information about the memory to be allocated may specifically be the memory size required by the target virtual machine.

In a specific implementation manner provided in this specification, taking virtual machine m as an example, a memory allocation request for virtual machine m is received, and the memory allocation request carries 2G of memory information to be allocated for virtual machine m.

S1004. Obtain the number of free memory blocks and attribute information of the free memory blocks.

The number of free memory blocks specifically refers to how many memory blocks are completely free in the current use of the host machine, that is, the number of unused memory blocks of the host machine. The free memory block attribute information specifically refers to the memory block capacity, physical address range, and the like of each unused free memory block.

In a specific implementation manner provided in this specification, following the above example, the number of free memory blocks obtained from the host machine is 3, and the size of each memory block is 512M. The physical address range of memory block 1 is add1024-add1536, the physical address range of memory block 2 is add1536-add2048, and the physical address range of memory block 3 is add2048-add2560.

S1006. Allocate a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block.

The first type of free memory block specifically refers to a whole continuous memory block allocated for the target virtual machine. The first type of free memory block needs to be determined according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory.

Specifically, allocating a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block includes:

Determine the number of pre-allocated free memory blocks according to the memory information to be allocated and the attribute information of the free memory blocks;

Allocating a first type of free memory block to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks.

In practical applications, the number of pre-allocated free memory blocks is determined according to the information of the memory to be allocated and the attribute information of the free memory blocks. In a host, the specifications of the memory are the same, and the attribute information of the memory blocks is also the same. For example, for a For 8G memory, there are 8 memory blocks, and the size of each memory block is 1G; for a 4G memory, there are 8 memory blocks, and the size of each memory block is 512M.

After determining the attribute information of the free memory block, combined with the memory information to be allocated, the number of pre-allocated free memory blocks can be determined. If the memory information to be allocated is 4G, and the attribute information of the free memory block is 1G, then the number of pre-allocated free memory blocks is 4; If the memory information to be allocated is 3G, and the attribute information of the free memory block is 512M, then the number of pre-allocated free memory blocks is 6.

After the number of pre-allocated free memory blocks is determined, the first type of free memory blocks can be allocated to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks. Specifically, allocating a first type of free memory block to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks includes:

When the number of pre-allocated free memory blocks is less than or equal to the number of free memory blocks, allocate a first type of free memory blocks to the target virtual machine according to the number of pre-allocated free memory blocks;

If the number of pre-allocated free memory blocks is greater than the number of free memory blocks, allocate a first type of free memory blocks to the target virtual machine according to the number of free memory blocks.

In practical applications, the target virtual machine is allocated the first type of free memory blocks according to the relationship between the number of pre-allocated free memory blocks and the number of free memory blocks. When the number of pre-allocated free memory blocks is less than or equal to the number of free memory blocks In the case of , allocate the first type of free memory blocks to the target virtual machine according to the number of pre-allocated free memory blocks; otherwise, allocate the first type of free memory blocks to the target virtual machine according to the number of free memory blocks.

For example, the number of pre-allocated free memory blocks is 4, and the number of free memory blocks is 5, then the number of free memory blocks of the first type is 4, and 4 free memory blocks of the first type are allocated for the target virtual machine. If the number of pre-allocated free memory blocks is 6 and the number of free memory blocks is 5, then the number of free memory blocks of the first type is 5, and five free memory blocks of the first type are allocated to the target virtual machine.

In practical applications, it usually occurs that all the memory required by the target virtual machine cannot be allocated by free memory blocks, or the number of free memory blocks is insufficient. At this time, it is also necessary to allocate non-free memory blocks for the target virtual machine. Specifically, after allocating a first type of free memory block to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks, the method further includes:

determining unallocated memory information according to the to-be-allocated memory information and the first type of free memory block;

If the unallocated memory information is not empty, allocate a second type of memory block to the target virtual machine according to the unallocated memory information.

Among them, the unallocated memory information specifically refers to the remaining memory information after the memory information to be allocated has been allocated to the first type of free memory block. For example, the memory information to be allocated is 3.2G memory, and the attribute information of the memory block is 1G. There are 3 free memory blocks of the first type. At this time, the memory information that can be allocated is 3G, and the unallocated memory information is 0.2G; if the memory information to be allocated is 4G, the attribute information of the memory block is 1G, and the first type of free memory If there are 4 blocks, the virtual machine can occupy 4 free memory blocks of the first type. At this time, the unallocated memory information is empty; if the memory information to be allocated is 4G, the attribute information of the memory block is 1G, and the free memory blocks of the first type If there are 3 memory blocks, the unallocated memory of the virtual machine The information is 1G.

When the unallocated memory information is empty, it means that the memory resources required by the virtual machine have been allocated. There is no need to allocate additional memory blocks.

When the unallocated memory information is non-empty, it means that there is no free memory block in the host machine or all free memory blocks cannot be allocated to the virtual machine. Memory block. In the embodiments provided in this manual, the second type of memory block specifically refers to a memory block that has been allocated with other virtual machine memory. For example, the memory capacity of a certain memory block is 1G, and there is 512M of space Allocated to virtual machine 1, there is 512M free memory space, then this memory block is the second type of memory block.

In a specific implementation manner provided in this specification, after the unallocated memory information is determined, a memory block of the second type is allocated to the target virtual machine according to the unallocated memory information. If the unallocated memory information is 0.2G, and the memory block attribute information is 1G, allocating a complete free memory block for it will cause waste of resources. In order to reduce resource waste, you can allocate it in the second type of memory block that has already been used 0.2G of space for the virtual machine.

The memory management method provided in this specification receives a memory release request for a target virtual machine; determines at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determines the memory block corresponding to each memory block to be released Memory block attribute information; determine that the memory block attribute information is the first type of memory block to be released as the first type of memory block to be released; release the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released Frees memory for a memory block. The memory management method provided in this manual can quickly complete large-scale memory release and reset within milliseconds by adjusting the refresh configuration parameters of memory blocks, greatly reducing resource contention and avoiding the need for newly scheduled virtual machines. Preempting resources and affecting the startup of newly scheduled virtual machines. By modifying the memory release mechanism, for continuous and complete memory blocks, clearing can be quickly completed by adjusting the refresh configuration parameters.

In addition, when the virtual machine is allocated, a complete and continuous memory block is allocated first for the virtual machine to use, so that in the subsequent memory release, the refresh configuration parameters of the modified memory can be used to quickly clear the complete memory.

Referring to accompanying drawing 2, Fig. 2 shows a flow chart of a memory management method provided according to another embodiment of this specification, which specifically includes the following steps:

Step 202: Receive a memory allocation request for the target virtual machine, wherein the memory allocation request carries memory information to be allocated.

Step 204: Obtain the number of free memory blocks and attribute information of the free memory blocks.

Step 206: Allocate a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block.

Optionally, allocating a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block includes:

Optionally, allocating a first type of free memory block to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks includes:

In the case that the number of pre-allocated free memory blocks is less than or equal to the number of free memory blocks, according to the pre-allocation Allocating the first type of free memory blocks for the target virtual machine by configuring the number of free memory blocks;

Optionally, after allocating a first type of free memory block to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks, the method further includes:

It should be noted that the memory management method provided by the embodiment of this specification is consistent with the method of S1002-S1006 in the above-mentioned embodiment. For the specific explanation of steps 202 to 206, please refer to the details of steps S1002-S1006 in the previous embodiment. I won't repeat it here.

The memory management method provided by the embodiment of this specification is suitable for allocating complete and continuous memory blocks to the virtual machine first when allocating memory for the target virtual machine. It is convenient to use the refresh configuration parameters of the modified memory to clear the memory quickly and completely in the subsequent memory release.

The memory management method will be further described below by taking the application of the memory management method provided in this specification in allocating memory for a virtual machine M as an example in conjunction with FIG. 3 . Wherein, FIG. 3 shows a flowchart of a processing process of a memory management method provided by an embodiment of this specification, which specifically includes the following steps.

Step 302: The virtual machine is powered on.

In the embodiment provided in this specification, the virtual machine M is started in the host machine.

Step 304: apply for memory.

In the embodiment provided in this specification, the virtual machine M applies for memory from the host machine.

Step 306: Determine whether the memory exceeds the size of a single memory block.

In the embodiment provided in this specification, it is judged whether the requested memory size exceeds the size of a single memory block. If the requested memory size exceeds the size of a single memory block, step 308 is performed; otherwise, step 310 is performed.

Step 308: Determine a free memory block, and allocate memory.

In the embodiment provided in this specification, the free memory block is determined in the host computer, and the free memory block is allocated to the virtual machine M, and the unallocated memory size of the virtual machine is monitored at the same time. If the free memory block happens to be allocated to the virtual machine machine M, and there is no remaining memory, execute step 312, and the memory allocation is completed. If the virtual machine M has remaining unallocated memory and less than one memory block, step 310 is executed.

Step 310: Determine the memory block of the second type, and allocate memory.

In the embodiment provided in this specification, the memory block of the second type is determined and the free memory in the memory block of the second type is allocated to the virtual machine M.

Step 312: memory allocation is completed.

In the embodiment provided in this specification, when all the memory required by the virtual machine M is allocated, the memory allocation work of the virtual machine M is completed.

The memory management method provided by the embodiment of this manual is applicable to when allocating memory for the target virtual machine, the complete A contiguous block of memory is allocated for use by the virtual machine. It is convenient to use the refresh configuration parameters of the modified memory to clear the memory quickly and completely in the subsequent memory release.

The following describes the memory management method further by taking the application of the memory management method provided in this specification in releasing and clearing the memory of the virtual machine M as an example in conjunction with FIG. 4 . Wherein, FIG. 4 shows a flowchart of a processing process of a memory management method provided by an embodiment of this specification, which specifically includes the following steps.

Step 402: Shut down the virtual machine.

In the embodiment provided in this specification, the virtual machine M is shut down in the host machine.

Step 404: apply for memory release.

In the embodiment provided in this specification, an application is made to release the memory occupied by the virtual machine M.

Step 406: Determine whether it is the first type of memory block to be released.

In the embodiment provided in this specification, it is judged whether the memory block occupied by the virtual machine M is the first type of memory block to be released. If yes, execute steps 410-414; if not, execute step 408.

Step 408: CPU is cleared.

In the embodiment provided in this specification, for the second type of memory block to be released, the process of memory release and clearing is completed by using the CPU to write 0 into the memory address.

Step 410: calling the PASR memory controller driver to adjust the refresh configuration parameter of the first type of memory block to be released to disabled.

In the embodiment provided in this manual, the driver of the PASR memory controller is called to adjust the refresh configuration parameter of the first type of memory block to be released to disabled, and count the disabled duration.

Step 412: Wait for T milliseconds.

In the embodiment provided in this specification, T milliseconds is the duration for storing data in the memory block, and when T milliseconds are exceeded, the data stored in the memory block will be lost.

Step 414: Call the PASR memory controller driver to adjust the refresh configuration parameter of the first type of memory block to be released to enable.

In the embodiment provided in this specification, after the disable time exceeds T milliseconds, the PASR memory controller driver is called to adjust the refresh configuration parameter of the first type of memory block to be released to enable.

Step 416: memory release is completed.

In the embodiment provided in this specification, so far, the memory release of the virtual machine M is completed.

The memory management method provided in this specification receives a memory release request for a target virtual machine; determines at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determines the memory block corresponding to each memory block to be released Memory block attribute information; determine that the memory block attribute information is the first type of memory block to be released as the first type of memory block to be released; release the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released Frees memory for a memory block. The memory management method provided in this manual can quickly complete large-scale memory release and reset within milliseconds by adjusting the refresh configuration parameters of memory blocks, which greatly reduces the competition for resources and avoids Preempting resources and affecting the startup of newly scheduled virtual machines. By modifying the memory release mechanism, for continuous and complete memory blocks, clearing can be quickly completed by adjusting the refresh configuration parameters.

Corresponding to the foregoing method embodiments, this specification also provides an embodiment of a memory management device. FIG. 5 shows a schematic structural diagram of a memory management device provided by an embodiment of this specification. As shown in Figure 5, the device includes:

The receiving module 502 is configured to receive a memory release request for the target virtual machine;

The response module 504 is configured to determine at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determine memory block attribute information corresponding to each memory block to be released;

The determining module 506 is configured to determine that the memory block whose attribute information is the first type to be released is the first type of memory block to be released, wherein the first type of memory block to be released is a complete memory block;

The release module 508 is configured to release the memory of the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released.

Optionally, the release module 508 is further configured to:

Optionally, the response module 504 is further configured to:

Optionally, the determining module 506 is further configured to:

Optionally, the device also includes:

The determination module 506 is further configured to determine that the memory block attribute information is that the memory block to be released of the second type is the memory block to be released of the second type, wherein the memory block to be released of the second type is a shared memory block;

The address obtaining module is configured to obtain the to-be-released physical address range corresponding to each second type of memory block to be released;

The clearing module is configured to clear the memory of each physical address range to be released.

Optionally, the device also includes:

An allocation module configured to allocate a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block.

Optionally, the allocation module is further configured to:

The memory management device provided in this specification receives a memory release request for a target virtual machine; determines at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determines the memory block corresponding to each memory block to be released Memory block attribute information; determine that the memory block attribute information is the first type of memory block to be released as the first type of memory block to be released; release the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released Frees memory for a memory block. The memory management device provided in this manual can quickly complete large-scale memory release and reset within milliseconds by adjusting the refresh configuration parameters of memory blocks, which greatly reduces the competition for resources and avoids Preempting resources and affecting the startup of newly scheduled virtual machines. By modifying the memory release mechanism, for continuous and complete memory blocks, clearing can be quickly completed by adjusting the refresh configuration parameters.

The foregoing is a schematic solution of a memory management device in this embodiment. It should be noted that the technical solution of the memory management device and the above-mentioned technical solution of the memory management method belong to the same idea, and details of the technical solution of the memory management device that are not described in detail can be found in the description of the technical solution of the above-mentioned memory management method .

Corresponding to the foregoing method embodiments, this specification also provides an embodiment of a memory management device. FIG. 6 shows a schematic structural diagram of a memory management device provided by another embodiment of this specification. As shown in Figure 6, the device includes:

The allocation request receiving module 602 is configured to receive a memory allocation request for the target virtual machine, wherein the memory allocation request carries memory information to be allocated;

The information obtaining module 604 is configured to obtain the number of free memory blocks and attribute information of free memory blocks;

The allocation module 606 is configured to allocate a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block.

Optionally, the allocation module 606 is further configured to:

determining unallocated memory information according to the to-be-allocated memory information and the first type of free memory blocks;

The memory management device provided by the embodiment of this specification is suitable for allocating complete and continuous memory blocks to the virtual machine with priority when allocating memory for the target virtual machine. It is convenient to use the refresh configuration parameters of the modified memory to clear the memory quickly and completely in the subsequent memory release.

FIG. 7 shows a structural block diagram of a computing device 700 provided according to an embodiment of this specification. Components of the computing device 700 include, but are not limited to, memory 710 and processor 720 . The processor 720 is connected to the memory 710 through the bus 730, and the database 750 is used for saving data.

Computing device 700 also includes an access device 740 that enables computing device 700 to communicate via one or more networks 760 . Examples of these networks include the Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or a combination of communication networks such as the Internet. Access device 740 may include one or more of any type of network interface (e.g., a network interface card (NIC)), wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, Worldwide Interoperability for Microwave Access ( Wi-MAX) interface, Ethernet interface, Universal Serial Bus (USB) interface, cellular network interface, Bluetooth interface, Near Field Communication (NFC) interface, etc.

In an embodiment of the present specification, the above-mentioned components of the computing device 700 and other components not shown in FIG. 7 may also be connected to each other, for example, through a bus. It should be understood that the structural block diagram of the computing device shown in FIG. 7 is only for the purpose of illustration, rather than limiting the scope of this description. Those skilled in the art can add or replace other components as needed.

Computing device 700 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, notebook computers, netbooks, etc.), mobile telephones (e.g., smartphones), ), wearable computing devices (eg, smart watches, smart glasses, etc.), or other types of mobile devices, or stationary computing devices such as desktop computers or PCs. Computing device 700 may also be a mobile or stationary server.

Wherein, the processor 720 is used to execute the following computer-executable instructions, which are executed by the processor During execution, the steps of the above data processing method are realized. The foregoing is a schematic solution of a computing device in this embodiment. It should be noted that the technical solution of the computing device and the above-mentioned technical solution of the memory management method belong to the same concept, and details not described in detail in the technical solution of the computing device can refer to the description of the technical solution of the above-mentioned memory management method.

An embodiment of the present specification also provides a computer-readable storage medium, which stores computer-executable instructions, and implements the steps of the memory management method above when the computer-executable instructions are executed by a processor.

The foregoing is a schematic solution of a computer-readable storage medium in this embodiment. It should be noted that the technical solution of the storage medium and the above-mentioned technical solution of the memory management method belong to the same idea, and details not described in detail in the technical solution of the storage medium can refer to the description of the technical solution of the above-mentioned memory management method.

An embodiment of the present specification further provides a computer program, wherein, when the computer program is executed in a computer, the computer is caused to execute the steps of the memory management method above.

The foregoing is a schematic solution of a computer program in this embodiment. It should be noted that the technical solution of the computer program and the above-mentioned technical solution of the memory management method belong to the same idea, and details not described in detail in the technical solution of the computer program can refer to the description of the technical solution of the above-mentioned memory management method.

The foregoing describes specific embodiments of this specification. Other implementations are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. Multitasking and parallel processing are also possible or may be advantageous in certain embodiments.

The computer instructions include computer program code, which may be in source code form, object code form, executable file or some intermediate form or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, computer-readable media Excludes electrical carrier signals and telecommunication signals.

It should be noted that, for the sake of simplicity of description, the aforementioned method embodiments are expressed as a series of action combinations, but those skilled in the art should know that the embodiments of this specification are not limited by the described action sequences. Because according to the embodiment of the present specification, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of the specification.

In the above-mentioned embodiments, the descriptions of each embodiment have their own emphases, and for parts not described in detail in a certain embodiment, reference may be made to relevant descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are only for helping to explain the present specification. Alternative embodiments are not exhaustive in all detail, nor are the inventions limited to specific implementations described. Obviously, many modifications and changes can be made according to the contents of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that those skilled in the art can well understand and use this specification. This specification is to be limited only by the claims, along with their full scope and equivalents.

Claims

A memory management method, comprising:

Receive a memory release request for the target virtual machine;

Determining at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determining memory block attribute information corresponding to each memory block to be released;

Determining that the memory block attribute information is the first type of memory block to be released is the first type of memory block to be released, wherein the first type of memory block to be released is a complete memory block;

The memory of the first type of memory block to be released is released by adjusting the refresh configuration parameters of the first type of memory block to be released.
The memory management method according to claim 1, releasing the memory of the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released, comprising:

Setting the refresh configuration parameter of the first type of memory block to be released to disabled, and counting the duration of the disabled;

When the disabled duration is greater than a preset threshold, the refresh configuration parameter of the first type of memory block to be released is set to enable, wherein the preset threshold is the data storage duration of the memory block to save data.
The memory management method according to claim 1, determining at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, comprising:

determining a virtual machine physical address range corresponding to the target virtual machine in response to the memory release request;

Determine the memory block corresponding to the physical address range of the virtual machine as the memory block to be released.
The memory management method according to claim 3, determining the memory block attribute information corresponding to each memory block to be released, comprising:

determining a target memory block to be released in the at least one memory block to be released;

Determine the physical address range of the memory block of the target memory block to be released;

Judging whether there is a memory block physical address in the physical address range of the memory block that does not belong to the physical address range of the virtual machine;

If not, then determine that the memory block attribute information of the target memory block to be released is the first type;

If yes, determine that the memory block attribute information of the target memory block to be released is the second type.
The memory management method according to claim 1, said method further comprising:

Determining that the memory block attribute information is that the second type of memory block to be released is the second type of memory block to be released, wherein the second type of memory block to be released is a shared memory block;

Obtain the to-be-released physical address range corresponding to each second type of memory block to be released;

Clear the memory of each physical address range to be released.
The memory management method according to claim 1, before receiving the memory release request for the target virtual machine, the method further comprises:

receiving a memory allocation request for the target virtual machine, wherein the memory allocation request carries memory information to be allocated;

Obtain the number of free memory blocks and attribute information of free memory blocks;

Allocating a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block.
A memory management method, comprising:

receiving a memory allocation request for the target virtual machine, wherein the memory allocation request carries memory information to be allocated;

Obtain the number of free memory blocks and attribute information of free memory blocks;

Allocating a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block, wherein the first type of free memory block is a complete memory block .
The memory management method according to claim 7, allocating a first type of free memory block for the target virtual machine according to the memory information to be allocated, the number of free memory blocks and the attribute information of the free memory block, comprising:

Determine the number of pre-allocated free memory blocks according to the memory information to be allocated and the attribute information of the free memory blocks;

Allocating a first type of free memory block to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks.
The memory management method according to claim 8, allocating a first type of free memory block for the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks, comprising:

When the number of pre-allocated free memory blocks is less than or equal to the number of free memory blocks, allocate a first type of free memory blocks to the target virtual machine according to the number of pre-allocated free memory blocks;

If the number of pre-allocated free memory blocks is greater than the number of free memory blocks, allocate a first type of free memory blocks to the target virtual machine according to the number of free memory blocks.
The memory management method according to claim 8, after allocating a first type of free memory block to the target virtual machine according to the number of pre-allocated free memory blocks and the number of free memory blocks, the method further comprises:

determining unallocated memory information according to the to-be-allocated memory information and the first type of free memory blocks;

If the unallocated memory information is not empty, allocate a second type of memory block to the target virtual machine according to the unallocated memory information.
A memory management device, comprising:

a receiving module configured to receive a memory release request for the target virtual machine;

A response module configured to determine at least one memory block to be released corresponding to the target virtual machine in response to the memory release request, and determine memory block attribute information corresponding to each memory block to be released;

The determining module is configured to determine that the attribute information of the memory block is that the first type of memory block to be released is the first type of memory block to be released, wherein the first type of memory block to be released is a complete memory block;

The release module is configured to release the memory of the first type of memory block to be released by adjusting the refresh configuration parameters of the first type of memory block to be released.
A memory management device, comprising:

An allocation request receiving module configured to receive a memory allocation request for the target virtual machine, wherein the memory allocation request carries memory information to be allocated;

An information acquisition module configured to acquire the number of free memory blocks and attribute information of the free memory blocks;

An allocation module configured to allocate a first type of free memory block to the target virtual machine according to the memory information to be allocated, the number of free memory blocks, and the attribute information of the free memory block, wherein the first type of free memory block A memory block is a complete memory block.
A computing device comprising:

memory and processor;

The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the memory described in any one of claims 1 to 6 or 7 to 10 is implemented. The steps of the management method.
A computer-readable storage medium, which stores computer-executable instructions, and implements the steps of the memory management method described in any one of claims 1-6 or 7-10 when the computer-executable instructions are executed by a processor.