Data distribution method and system for hyper-converged system

A data allocation and hyper-convergence technology, applied in the field of hyper-convergence systems, can solve problems such as limited physical network card bandwidth, increased network resource consumption, and reduced reliability, achieving efficient and reliable data allocation, safe and reliable storage placement, and reduced processing delays Effect

Active Publication Date: 2022-03-25
北京志凌海纳科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Access to these data blocks needs to go through a cross-node network. The performance will increase the delay due to the increase of network transit links, and the bandwidth will also be limited by the bandwidth of the physical network card, which will increase the consumption of external network resources (switches).
[0007] (2) Reduced reliability
[0008] To sum up, the current existing technology has the objective shortcomings of low system performance and low reliability, and it is urgent to invent an efficient and reliable data distribution method and system for hyper-converged systems

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
  • Data distribution method and system for hyper-converged system
  • Data distribution method and system for hyper-converged system
  • Data distribution method and system for hyper-converged system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] refer to figure 2 , a data distribution method for a hyper-converged system includes:

[0065] Step S101, splitting the data of the virtual disk of the virtual machine to form several data blocks, and each data block has multiple copies of the data block.

[0066] Step S102, in the hyper-converged physical server cluster, perform localized spatial location allocation for each data block copy according to the running status of the virtual machine and the cluster data status, so that all data blocks on the same virtual disk have a copy of the data block Placed on the same physical server as the virtual machine. Among them, the virtual machine running status includes CPU usage, memory usage, disk health status, storage system IO pressure, and remaining disk space; cluster data status includes new physical service addition, physical server failure, disk failure, and virtual machine migration.

[0067] Step S103, perform localized allocation on other data block copies exc...

Embodiment 2

[0091] Figure 10 The schematic diagram of the system structure based on the hyper-converged infrastructure provided by the embodiment of the present invention, combined with this embodiment, can better understand the implementation of the method in Embodiment 1.

[0092] The system may include: physical servers, the disks on the physical servers are used to finally store data, the data link service running on each physical server is used to process the data access requests of virtual machines, and the data management service on each physical server Used to manage disks, such as the final storage location of data on disks, each physical server will run some virtual machines, and they access the virtual disks provided by the hyper-converged storage system through data link services. At the same time, a replica allocation service is also deployed in the cluster to determine which physical servers in the cluster the split data blocks of each virtual disk should be placed on.

[...

Embodiment 3

[0101] refer to Figure 11 , a data distribution system of a hyper-converged system includes:

[0102] The split unit 10 is used to split the data of the virtual machine virtual disk to form several data blocks, and each data block has a plurality of data block copies;

[0103] The first allocation unit 20 is configured to perform localized space allocation for each data block copy according to the running state of the virtual machine and the cluster data state in the hyper-converged physical server cluster, so that all data blocks on the same virtual disk are There is a copy of the data block placed on the physical server where the virtual machine resides;

[0104] The second allocating unit 30 is configured to perform localized allocation on other data block copies except the localized spatial position allocation, so that all data block copies on the same virtual disk can be gathered in order according to rules.

[0105] It can be understood that the above data distributio...

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 embodiment of the present invention discloses a data distribution method and system of a hyper-converged system, including: splitting the data of a virtual disk of a virtual machine to form several data blocks, each data block having multiple data block copies; According to the virtual machine running status and cluster data status, each data block copy is assigned a local space location, so that all data blocks on the same virtual disk have a copy of the data block placed on the physical server where the virtual machine is located. ; Perform localized allocation on other data block copies except localized space location allocation, so that all data block copies on the same virtual disk can be gathered in order according to rules. The present invention reduces request processing delay, increases data bandwidth, makes data storage and placement safer and more reliable, and realizes efficient and reliable data distribution of hyper-converged clusters.

Description

technical field [0001] The present invention relates to the technical field of hyper-converged systems, in particular to a data distribution method and system of a hyper-converged system. Background technique [0002] Hyper-converged infrastructure is a technical architecture that integrates resources such as computing, network, and storage as infrastructure, selects, combines, and customizes according to specific business system requirements, and facilitates data center construction and business system deployment. Generally, software virtualization technology (including virtualization of computing, network, storage, security, etc.) is integrated into unit nodes (x86 servers), and each unit node is aggregated through the network to realize modular seamless horizontal expansion and build a unified resource pool . [0003] In a hyperconverged infrastructure, the computing load (applications / virtual machines) and associated data reside on the same set of physical servers. Dif...

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 Patents(China)
IPC IPC(8): G06F9/455G06F9/50
CPCG06F9/45558G06F9/5077G06F2009/4557G06F2009/45575G06F2009/45579G06F2009/45583
Inventor 周于添
Owner 北京志凌海纳科技有限公司
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap