Resource scheduling allocation method, computer system and super-fusion architecture system

An allocation method and resource scheduling technology, which are applied in the fields of resource scheduling and allocation methods, computer systems and hyper-converged architecture systems, can solve problems such as large functional limitations such as migration and hot add, affecting the overall performance of the system, and large loss of memory access performance. , to solve the problem of complex resource contention, ensure available resources, and reduce resource contention

Active Publication Date: 2018-06-22
武汉泽塔云科技股份有限公司
View PDF8 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there are mainly OpenStack solutions and VMware solutions for virtualized NUMA management, but the OpenStack solution completely relies on the user-specified NUMA structure and mapping strategy, and the VMware solution adopts the "NUMA affinity" strategy, which is not strictly bound. Both the performance of the system and the system itself are not good, and there are great restrictions on functions such as migration and hot addition
To sum up, these two solutions cannot be flexibly configured to build a reasonable NUMA structure for the virtual machine, resulting in a large loss of memory access performance and affecting the overall performance of the system

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Resource scheduling allocation method, computer system and super-fusion architecture system
  • Resource scheduling allocation method, computer system and super-fusion architecture system
  • Resource scheduling allocation method, computer system and super-fusion architecture system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0084] A resource scheduling allocation method, such as figure 1 shown, including the following steps:

[0085] 101. Determine the overall configuration requirements of the virtual NUMA structure corresponding to the virtual machine. The virtual NUMA structure includes several virtual NUMA nodes. The virtual NUMA nodes include virtual CPU cores and virtual memory. The overall configuration requirements include sub-configuration requirements of virtual NUMA nodes, sub-configuration The requirements include: a first configuration requirement for virtual CPU cores and a second configuration requirement for virtual memory.

[0086] 102. Determine the overall resource availability of the physical NUMA structure. The physical NUMA structure includes a number of physical NUMA nodes. The physical NUMA nodes include physical CPU cores and physical memory. The overall resource availability includes the availability of sub-resources of the physical NUMA nodes. The availability of ...

Embodiment 2

[0104] The difference between this embodiment and other embodiments is that in this embodiment, the method for resource scheduling and allocation further includes: reserving resources that will not be occupied by non-virtual machine loads for the virtual NUMA structure in the physical NUMA structure, That is, clarify the resources occupied by the resource contention of storage, network and computing (non-virtual machine internal) functions, accurately determine the total amount of virtual resources available at system initialization, and ensure that these virtual resources will not be occupied by non-virtual machine loads , so as to reduce resource contention and ensure the available resources of the virtual machine.

[0105] In this embodiment, the resource scheduling allocation method further includes: when the overall resource availability indicates that the physical NUMA structure cannot meet the requirements of the virtual NUMA structure, incorporating a new physical N...

Embodiment 3

[0108] This embodiment describes a detailed embodiment of the resource scheduling allocation method, which mainly includes the following parts:

[0109] 1) A method for constructing a reasonable virtual NUMA structure when creating a virtual machine;

[0110] 2) The resource mapping relationship, memory allocation strategy and CPU binding strategy when the virtual machine is powered on;

[0111] 3) Resource mapping relationship, memory allocation strategy and CPU binding strategy when hot adding CPU and memory;

[0112] 4) Scheduling algorithm during cold / hot migration, virtual NUMA structure reconstruction strategy during cold migration / resizing, resource remapping method during hot migration, including memory allocation strategy adjustment and CPU rebinding;

[0113] 5) Optimization method for resource contention under the hyper-converged architecture.

[0114] First, the method of constructing a reasonable virtual NUMA structure when creating a virtual machi...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a resource scheduling allocation method, a computer system and a super-fusion architecture system. The resource scheduling allocation method comprises the steps of determiningan overall configuration demand of a virtual NUMA structure corresponding to a virtual machine, wherein the overall configuration demand comprises sub-configuration demands of a virtual NUMA node, and the sub-configuration demands include a first configuration demand for a virtual CPU core and a second configuration demand for a virtual memory; determining overall resource availability of a physical NUMA structure, wherein the overall resource availability comprises sub-resource availability of a physical NUMA node, and the sub-resource availability includes first resource availability for aphysical CPU core and second resource availability for a physical memory; and based on the overall configuration demand and the overall resource availability, establishing a corresponding relationshipbetween the virtual NUMA node and the physical NUMA node. The resource scheduling allocation method realizes mapping and binding of the virtual NUMA structure and the physical NUMA structure, so thatphysical resources are reasonably allocated.

Description

technical field [0001] The present invention relates to the field of communication technology, and more specifically, it relates to a resource scheduling and allocation method, a computer system and a hyper-converged architecture system. Background technique [0002] Hyper-converged architecture refers to the use of software methods to realize and integrate computing, network, and storage functions on servers (especially x86 servers) to form a modular deployment unit, and interconnect the network through switching (constructed by standard commercial switches, etc.) Realize a cluster that can scale horizontally. [0003] At present, the multi-processor computer system structure is composed of a non-uniform memory access (Non Uniform Memory Access Architecture, NUMA) architecture. Under the NUMA architecture, the time for the processor to access the memory depends on the location of the memory. Usually, the processor accesses its local memory faster than the non-local memory....

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/50
CPCG06F9/5077
Inventor 黄扬
Owner 武汉泽塔云科技股份有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products