CN103888492A - Memory management system and method - Google Patents

Abstract

The invention provides a memory management system applied to a cloud server. The memory management system comprises a first receiving module used for receiving a request of a cloud client for requesting the cloud server to designate a physical memory space of the first capacity, a memory distribution module used for designating a fixed address space in the physical memory of the cloud server for the client end for satisfying the first capacity, a second receiving module used for receiving a request of the cloud client for requesting the cloud server to distribute the memory of the second capacity when the cloud client operates an application program, and an address conversion module used for distributing the virtual memory of the second capacity for the application program of the cloud client and converting the address space of the virtual memory into the fixed address space of the physical memory. The invention further provides a memory management method.

Description

Memory management system and method

technical field

The invention relates to a system for providing cloud services, in particular to a system and method for managing the memory of a cloud server.

Background technique

With the development of cloud technology, more and more network services are transferred to cloud services. For customers, cloud services can be divided into three types: public cloud, private cloud and hybrid cloud. Among them, the private cloud server can allow small customers to jointly rent the same cloud server, and different cloud clients can share the services provided by the private cloud server, such as physical memory. When different cloud clients request the private cloud server to allocate memory to run applications, the private cloud server provides each cloud client with a virtual memory address space of corresponding capacity, and these virtual memory spaces are mapped to the address space of the physical memory. However, this mapping relationship is dynamic, and applications of different cloud clients may cross-use the same address space of physical memory, which may not only cause the application to run slowly, but may also bury data security risks.

Contents of the invention

In view of the above, it is necessary to provide a memory management system and method, which can allocate physical memory with a fixed address space for the client of the private cloud, prevent different client applications from accessing the physical memory of the same address space, and improve the performance of the client application. Speed of operation and data security.

A memory management system applied to cloud servers. The system includes: a first receiving module, configured to receive a request from a cloud client to request a cloud server to specify a first-capacity physical memory space; The fixed address space of the capacity; The second receiving module is used for receiving the request that the cloud server allocates the memory of the second capacity when the cloud client runs the application program; and the address translation module is used for the application program allocation of the cloud client. virtual memory of the second capacity, and convert the address space of the virtual memory into the fixed address space of the physical memory.

A memory management method, the method comprising: (A) receiving a request from a cloud client requesting a cloud server to specify a first-capacity physical memory space; (B) specifying that the cloud client satisfies the first capacity in the physical memory of the cloud server (C) receiving a request from the cloud server to allocate memory of a second capacity when the cloud client runs an application; and (D) allocating virtual memory of the second capacity to the application program of the cloud client, and The address space of the virtual memory is translated into said fixed address space of the physical memory.

Compared with the prior art, the memory management system and method provided by the present invention can allocate physical memory with a fixed address space for the cloud client, prevent different client applications from accessing the same address space in the physical memory of the cloud server, and improve customer The running speed and data security of terminal applications.

Description of drawings

FIG. 1 is an application environment diagram of a preferred embodiment of the memory management system of the present invention.

Fig. 2 is a functional block diagram of a preferred embodiment of the memory management system of the present invention.

FIG. 3 is a flowchart of a preferred embodiment of the memory management method of the present invention.

Description of main component symbols

private cloud server 1 network 2 cloud client 3 memory management system 10 first receiving module 11 memory allocation module 12 Second receiving module 13 address translation module 14 physical memory 20 memory 30

processor 40

The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings.

Detailed ways

Referring to FIG. 1 , it is an application environment diagram of a preferred embodiment of the memory management system 10 of the present invention. The memory management system 10 is applied to a private cloud server 1, and multiple cloud clients 3 are connected to the private cloud server 1 through a network 2 to share services provided by the private cloud server. The memory management system 10 allocates a fixed address space in the physical memory 20 to each cloud client 3 , preventing applications of different cloud clients 3 from accessing the same address space in the physical memory 20 .

Referring to FIG. 2 , it is a functional block diagram of a preferred embodiment of the memory management system 10 of the present invention. The memory management system 10 has a first receiving module 11 , a memory allocation module 12 , a second receiving module 13 and an address translation module 14 . The modules 11-14 include computer programming instructions, which are stored in the preset address space of the memory 30 of the private cloud server 1, and the processor 40 of the private cloud server 1 executes these computer programming instructions to provide the memory management system 10 of the above functions. The specific functions of the modules 11-14 will be described below in conjunction with FIG. 3 .

Referring to FIG. 3 , it is a flowchart of a preferred embodiment of the memory management method of the present invention.

Step S10, the first receiving module 11 receives a request from the cloud client 3 requesting the private cloud server 1 to specify a physical memory space of the first capacity. For example, the first receiving module 11 receives the request from the first cloud client 3 requesting the private cloud server 1 to specify 500MB of physical memory space, and receives the request from the second cloud client 3 requesting the private cloud server 1 to specify 400MB of physical memory space.

Step S20 , the memory allocation module 12 assigns a fixed address space satisfying the first capacity in the physical memory 20 for the cloud client 3 . For example, the memory allocation module 12 allocates the address space 0x00001000~0x******** of 500MB from the physical address 0x00001000 in the physical memory 20 for the first cloud client 3, and allocates the address space 0x00001000~0x******** for the second cloud client 3 in the physical memory 20. The address space of 400MB is allocated from the physical address 0x****** in the physical memory 20 .

In step S30, the second receiving module 13 receives a request from the cloud client 3 to allocate memory of the second capacity when running the application program.

Step S40 , the address conversion module 14 allocates the virtual memory of the second capacity to the application program of the cloud client 3 , and converts the address space of the virtual memory into the fixed address space of the physical memory 20 . For example, assuming that the first cloud client 3 runs application A and requests to allocate 480MB of memory space, and the second cloud client 3 runs application B and requests to allocate 450MB of memory space, then the address translation module 14 allocates 480MB for application A The virtual memory space of 480MB is converted into the fixed address space 0x00001000~0x******** of physical memory 20. The address translation module 14 allocates 450MB of virtual memory space for the application program B, and converts the 450MB of virtual memory space into the address space of 400MB allocated from the physical address 0x****** in the physical memory 20 .

In step S50, the address conversion module 14 judges whether the fixed address space of the physical memory 20 can meet the requirement of the virtual memory. If the requirement of the virtual memory can be met, the application program of the cloud client 3 is run, and then the process enters step S70. For example, the application A of the first cloud client 3 needs 480MB of memory space, while the first cloud client 3 specifies a fixed address space of 500MB in the physical memory 20, which can meet the requirements of the application A. If the fixed address space of the physical memory 20 cannot meet the requirements of the virtual memory, step S60 is executed. For example, the second cloud client 3 requires 450MB of memory space to run the application program B, while the fixed address space specified in the physical memory 20 of the second cloud client 3 is only 400MB, which is still 50MB short.

In step S60, the address translation module 14 divides part of the address space from the memory 30 to supplement the virtual memory. For example, the address translation module 14 divides an address space of 50MB from the memory 30 to supplement the virtual memory. The memory 30 may include a hard disk, a flash disk, and the like.

In step S70, the memory is released after the application program of the cloud client 3 finishes running. For example, after application A finishes running, it releases the 480MB memory space it occupies in physical memory 20; 50MB of address space. In other embodiments, the memory management method may not include step S70.

It should be noted that the memory management method provided by the present invention can also be applied to public cloud and hybrid cloud.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention without limitation, although the present invention has been described in detail with reference to the above preferred embodiments, those of ordinary skill in the art should understand that the present invention can be The technical solution shall be modified or equivalently replaced without departing from the spirit and scope of the technical solution of the present invention.

Claims (6)

1. A memory management method, characterized in that the method comprises: The first receiving step: receiving a request from the cloud client to request the cloud server to specify the physical memory space of the first capacity; Memory allocation step: specifying a fixed address space satisfying the first capacity for the cloud client in the physical memory of the cloud server; The second receiving step: receiving a request from the cloud client to request the cloud server to allocate a second capacity memory when running the application program; and Address translation step: Allocating virtual memory of the second capacity to the application program of the cloud client, and converting the address space of the virtual memory into the fixed address space of the physical memory. 2. The memory management method according to claim 1, wherein the address translation step further comprises: when the fixed address space of the physical memory cannot meet the requirements of the virtual memory of the second capacity, Part of the address space is divided from the storage of the cloud server to supplement the virtual memory. 3. The memory management method according to claim 2, wherein the memory comprises a hard disk and a flash disk. 4. A memory management system, characterized in that the system comprises: The first receiving module is configured to receive a request from a cloud client to request a cloud server to specify a first-capacity physical memory space; A memory allocation module, configured to specify a fixed address space satisfying the first capacity for the cloud client in the physical memory of the cloud server; The second receiving module is used to receive a request for the cloud server to allocate a second capacity of memory when the cloud client runs the application program; and The address conversion module is configured to allocate the virtual memory of the second capacity to the application program of the cloud client, and convert the address space of the virtual memory into the fixed address space of the physical memory. 5. The memory management system according to claim 4, wherein the address translation module is further configured to: when the fixed address space of the physical memory cannot meet the requirements of the virtual memory of the second capacity , divide part of the address space from the storage of the cloud server to supplement the virtual memory. 6. The memory management system according to claim 5, wherein the memory comprises a hard disk and a flash disk.

Images