WO2023080339A1

WO2023080339A1 - Memory management device for virtual machine

Info

Publication number: WO2023080339A1
Application number: PCT/KR2022/000417
Authority: WO
Inventors: 엄상호; 류대수; 김재신
Original assignee: 주식회사 시스기어
Priority date: 2021-11-08
Filing date: 2022-01-11
Publication date: 2023-05-11
Also published as: KR20230067755A; TW202319924A

Abstract

Disclosed is a device for managing a virtual memory of a virtual machine composed of a plurality of computing devices that are physically independent and interconnected through a network. A memory management device according to the present invention is characterized by comprising: a virtual memory monitoring unit which monitors real-time usage of each of a plurality of guest memory areas; a local memory monitoring unit which monitors real-time usage and idle memory capacity of a local memory of each of a plurality of computing devices; and a virtual memory management unit which, when a real-time usage rate of a specific guest memory area is equal to or higher than a reference value, identifies at least one computing device that can be used as an idle memory and transmits same to a master node.

Description

Virtual machine memory management device

The present invention relates to a computing system, and more particularly, to a memory management device for a virtual machine for efficiently managing a memory of the virtual machine.

With the development of networking technology, a cloud system for sharing computing resources is being provided. A cloud system may be configured so that multiple users can share various physical infrastructures. The physical infrastructures may include one or more computing systems having processors, memory, storage, networking, and the like. Physical infrastructure must be used to implement or run the workloads required by the application. However, since the physical infrastructure must be shared within the cloud system, the cloud system manages the physical infrastructure through logical servers or virtual machines.

A virtual machine (VM) is a software implementation of a virtual computing environment rather than a physically existing computer, that is, a technology for emulating (virtualizing) a computer system. An operating system or application program can be installed and executed in a virtual machine. A virtual machine may be implemented through a plurality of computing devices that are physically independent and connected to each other through a predetermined communication network. Through this, users can use the same environment regardless of physical location.

In a virtualization system, for the convenience of users performing tasks using resources of servers connected through computer networking in the Internet and/or intranet environment, it must be available 24 hours a day and must be able to avoid service interruption. Due to the rapid increase in data-intensive workloads such as in-memory databases, data caching, bioinformics, and graph processing, the memory capacity required for cloud systems is increasing. . In order to provide a large-capacity cloud system in line with this trend, a large-capacity physical server infrastructure is required, but a lot of cost is required to build it. Therefore, there is a need to efficiently manage idle memory resources in each server while using a plurality of small-capacity servers instead of large-capacity physical servers and using memories distributed in different physical computing devices as a single memory.

An object of the present invention is to provide a memory management device for a virtual machine capable of more efficiently using memory resources of a plurality of servers connected through networking in a virtualization environment.

A memory management apparatus according to the present invention is a memory management apparatus for a virtual machine composed of a plurality of computing devices that are physically independent and interconnected through a network, wherein one of the plurality of computing devices is a master node and the others are agent nodes. At this time, the memory of the virtual machine created by the master node is composed of one virtual memory by allocating the local memories of the plurality of computing devices, and the virtual memory is divided into a plurality of guest memory areas, and each of the plurality of guests and the memory management device includes a virtual memory monitoring unit for monitoring real-time usage of each of the plurality of guest memory areas allocated to each of the plurality of guests, real-time usage of local memory of each of the plurality of computing devices, and A local memory monitoring unit for monitoring idle memory capacity, and at least one of the local memories of the plurality of computing devices that can be used as idle memory if the real-time usage rate of a specific guest memory area allocated to any one of the plurality of guests exceeds a reference value. and a virtual memory management unit for specifying a computing device of and transmitting information on the specific guest memory area and the specific computing device to use the idle memory to the master node.

The virtual memory monitoring unit may monitor real-time data transfer of a guest operating system in each guest memory area of the virtual memory created by the master node to calculate real-time usage in the guest memory area.

In addition, the agent nodes transmit data on the usage of each local memory and data on the capacity of idle memory that is not being used to the master node in real time, and the local memory monitoring unit monitors data between the master node and the agent nodes. It is characterized in that the real-time usage and idle memory capacity of the local memory of each of the plurality of computing devices is monitored based on the real-time data being communicated.

In addition, the memory management device may be provided in the computing device functioning as the master node. Also, the memory management device may be connected to the master node and the plurality of agent nodes through the network.

According to the present invention, a virtual memory of a virtual machine can be used more efficiently. In particular, when the memory area allocated to the guest using the virtual machine is insufficient, it is easy to additionally allocate the insufficient memory capacity in real time.

1 is a diagram schematically illustrating a virtualization system to which an apparatus for managing a memory of a virtual machine according to the present invention is applied.

2 is a block diagram illustrating the structure of a memory management device for a virtual machine according to the present invention.

Prior to the detailed description of the present invention, the terms or words used in this specification and claims described below should not be construed as being limited to a common or dictionary meaning, and the inventors should use their own invention in the best way. It should be interpreted as a meaning and concept corresponding to the technical idea of the present invention based on the principle that it can be properly defined as a concept of a term for explanation. Therefore, the embodiments described in this specification and the configurations shown in the drawings are only the most preferred embodiments of the present invention, and do not represent all of the technical ideas of the present invention, so various equivalents that can replace them at the time of the present application. It should be understood that there may be water and variations.

In the disclosure of the present invention, when a component is said to be "connected", "coupled" or "connected" to another component, this is not only a direct connection relationship, but also an indirect relationship between which another component exists. Connections may also be included. In addition, when a component "includes" or "has" another component, this means that it may further include another component without excluding other components unless otherwise stated. do. In addition, components that are distinguished from each other are intended to clearly explain each characteristic, and do not necessarily mean that the components are separated. That is, a plurality of components may be integrated to form a single hardware or software unit, or a single component may be distributed to form a plurality of hardware or software units. Therefore, even such integrated or distributed embodiments are included in the scope of the present invention, even if not separately mentioned.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. At this time, it should be noted that the same components in the accompanying drawings are indicated by the same reference numerals as much as possible. In addition, detailed descriptions of well-known functions and configurations that may obscure the gist of the present invention will be omitted.

First, FIG. 1 schematically illustrates a system to which an apparatus for managing a memory of a virtual machine according to the present invention is applied. The virtual machine (VM) may be created by any one of a plurality of computing devices (S1, S2, S3, ..., Sn) interconnected through the network 10. Each computing device can separately create a virtual machine independent of each other, and a specific computing device (S1) that created a specific virtual machine is referred to as a master node (MN), and the remaining computing devices (S2) connected to the master node (MN) , S3, ..., Sn) is called an agent node (AN). One computing device can function as a master node in a virtual machine created by itself, and at the same time function as an agent node in another virtual machine created by another computing device. In addition, one master node (MN) may create a plurality of virtual machines (VMs).

Furthermore, the master node (MN) creates a virtual machine using the computing resources of other agent nodes (AN) including itself. The master node (MN) creates a virtual machine (VM) using the computing resources of other agent nodes (AN) including itself. The master node (MN) is a plurality of computing devices (S1, S2, S3, …, Sn) computing resources may be allocated to virtual machines (VMs). These computing resources may include several types of physical resources. For example, physical resources include CPU, memory (eg, Random Access Memory (RAM)), storage (eg, hard disk or solid state drive), network I/O devices (eg, network interface cards), power devices (e.g., Power Bricks), cooling devices (e.g., fans or coolants), or other types of resources (e.g., network switch devices). .

On the other hand, the memory of the virtual machine (VM) created by the master node (MN) is composed of one virtual memory (100) by allocating local memories provided in a plurality of computing devices (S1, S2, S3, ..., Sn). do. In addition, the virtual memory 100 is provided to each of a plurality of guests by a size required for a user terminal (not shown; hereinafter 'guest') receiving services through a virtual machine to drive each operating system and application program. That is, the virtual memory 100 is divided into a plurality of guest memory areas (G1, G2, G3, ..., Gm) having an arbitrary size and allocated to each of the plurality of guests.

Depending on the activity status of each guest using the virtual machine (VM), the memory usage used by a specific guest may rapidly increase and become insufficient. there is. That is, for the management of virtual memory, as shown in FIG. 2, the memory management device 200 may be configured.

As shown in FIG. 2, the memory management device 200 of the virtual machine (VM) monitors the real-time usage of each of a plurality of guest memory areas (G1, G2, G3, ..., Gm) allocated to each of a plurality of guests. The virtual memory monitoring unit 210, the local memory monitoring unit 220 for monitoring the real-time usage amount of the local memory of each of the plurality of computing devices (S1, S2, S3, ..., Sn), and any one of the plurality of guests If the real-time usage rate of the allocated specific guest memory area is higher than the reference value, the virtual memory management unit 230 allocates the idle memory of at least one local memory among the plurality of computing devices (S1, S2, S3, ..., Sn) to the specific guest memory area. ).

The virtual memory monitoring unit 210 monitors real-time data transmission of a guest operating system (hereinafter referred to as a guest OS) in each guest memory area of the virtual memory 100 created by the master node (MN). For example, the virtual memory monitoring unit 210 directly contacts a guest operating system operated in each guest memory area (G1, G2, G3, ..., Gm) in a method such as SSH (Secure Shell) and echoes from each guest memory area. Receives data on real-time memory usage in this way. As a result, real-time usage in each guest memory area is calculated, and if the memory in use shows a continuous usage of more than a certain size (eg, 80%) compared to the size of the allocated guest memory area, it is determined that the area requires additional memory input. can

Next, the agent nodes (AN) transmit data on the amount of use of their local memory and the amount of idle memory that is not being used to the master node (MN) in real time. At this time, in order to store and display the data for each time slot in a file in consideration of the average load of the server, the storage space can be efficiently used while storing the data for each time slot in a round-robin manner. As one of preemptive scheduling, it is a method that does not prioritize between processes, but always in order by time unit (Time Quantum). Data collection methods include a poll method using SNMP (Simple Network Management Protocol), SSH (Secure Shell), TCP (Transmission Control Protocol), ICMP (Internet Control Message Protocol), IPMI (Intelligent Platform Management Interface), Telnet, You can use gRPC, HTTP2, etc. Based on the real-time data communicated between the master node (MN) and the agent node (AN), the local memory monitoring unit 220 monitors the local memory of each of the plurality of computing devices (S1, S2, S3, ..., Sn). Monitor real-time usage and idle memory capacity of

The virtual memory management unit 230, when it is confirmed by the virtual memory monitoring unit 210 that the memory usage in a specific guest memory area is equal to or greater than the reference value, the plurality of computing devices S1 and S2 checked by the local memory monitoring unit 220. , S3, ..., Sn) by specifying a computing device that can be used as an idle memory among local memories, and a specific guest memory area (at least one of G1 to Gm) to which memory is allocated and a specific computing device to use the idle memory (of S1 to Sn). at least one) to the master node (MN). Through this, the master node (MN) additionally allocates an idle memory surplus to a specific guest memory area through the hypervisor 20 .

The memory management device 200 of the above-described virtual machine (VM) may be directly installed in a computing device functioning as a master node (VM), or provided as a separate computing device and is connected to the master node (MN) and It can be connected to a plurality of agent nodes (AN).

The embodiments of the present disclosure are not intended to list all possible combinations, but are intended to explain representative aspects of the present disclosure, and matters described in various embodiments may be applied independently or in combination of two or more.

In addition, various embodiments of the present disclosure may be implemented by hardware, firmware, software, or a combination thereof. For hardware implementation, one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), It may be implemented by a processor (general processor), controller, microcontroller, microprocessor, or the like.

The scope of the present disclosure is software or machine-executable instructions (eg, operating systems, applications, firmware, programs, etc.) that cause operations according to various embodiments to be executed on a device or computer, and such software or instructions, etc. and a non-transitory computer-readable medium executable on a device or computer.

Claims

A memory management device for a virtual machine composed of a plurality of physically independent computing devices interconnected through a network,

When one of the plurality of computing devices is used as a master node and the others are used as agent nodes, the memory of the virtual machine created by the master node is allocated to the local memory of the plurality of computing devices to form one virtual memory. become,

The virtual memory is divided into a plurality of guest memory areas and allocated to each of a plurality of guests;

The memory management device,

a virtual memory monitoring unit for monitoring real-time usage of each of the plurality of guest memory areas allocated to each of the plurality of guests;

a local memory monitoring unit for monitoring real-time usage and idle memory capacity of the local memory of each of the plurality of computing devices;

If the real-time usage rate of a specific guest memory area allocated to any one of the plurality of guests exceeds the reference value, at least one computing device that can be used as an idle memory among the local memories of the plurality of computing devices is specified, and the specific guest memory area and a virtual memory management unit that transmits information on a specific computing device to use the idle memory to the master node.

Memory management unit of virtual machine.
According to claim 1,

The virtual memory monitoring unit monitors real-time data transmission of a guest operating system in each guest memory area of the virtual memory created by the master node to calculate real-time usage in the guest memory area.

Memory management unit of virtual machine.
According to claim 1,

The agent nodes transmit data on the amount of usage of each local memory and the amount of idle memory that is not being used to the master node in real time;

Characterized in that the local memory monitoring unit monitors real-time usage and idle memory capacity of the local memory of each of the plurality of computing devices based on real-time data communicated between the master node and the agent nodes.

Memory management unit of virtual machine.
According to claim 1,

Characterized in that the memory management device is provided in the computing device functioning as the master node.

Memory management unit of virtual machine.
According to claim 1,

Characterized in that the memory management device is connected to the master node and the plurality of agent nodes through the network.

Memory management unit of virtual machine.