Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A cross-node multi-link redundancy protection method and system

A redundant protection and multi-link technology, applied in the field of Multipath, can solve the problems of not being able to provide high-availability switching at a high level, and achieve the effect of ensuring continuity and improving economic benefits

Active Publication Date: 2021-06-18
HANGZHOU WOQU NETWORK TECH
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0012] Aiming at the prior art, the present invention provides a single-node single-link or single-node multi-link aggregation technology, which is not suitable for computing and storage separation architectures. In computing nodes, block devices from different storage nodes are linked Aggregation, which cannot provide high-availability switching and ensure I / O continuity, provides a cross-node multi-link redundancy protection method and 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
  • A cross-node multi-link redundancy protection method and system
  • A cross-node multi-link redundancy protection method and system
  • A cross-node multi-link redundancy protection method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] A cross-node multi-link redundancy protection method, comprising the steps of:

[0040] Step 1, block device aggregation, aggregate block devices of different links into a block device of one link, and accept I / O requests from applications;

[0041] Step 2: Judgment of the state of the block device. The state of the block device aggregated in step 1 is judged according to the I / O request in step 1. If there is a block device in the active state, the device is in the active state; otherwise, the I / O in step 1 is ended. O request.

[0042]By confirming the active state of the block device, it is judged whether it is necessary to continue issuing I / O requests. If there is no active block device, it is not necessary to issue I / O requests.

[0043] The method of aggregation of different link block devices in step 1 is the Multipath method.

[0044] The number of block devices is greater than or equal to 1; block devices include storage nodes and computing nodes.

[0045] ...

Embodiment 2

[0047] On the basis of embodiment 1, this embodiment also includes the following steps,

[0048] Active block device selection, select all active block devices according to the polling algorithm;

[0049] Confirm the request result, send the I / O request in step 1 to the selected active block device, and obtain the request result;

[0050] Judgment of the request result, if the request result is a confirmation value, then end the I / O request in step 1, otherwise continue to execute the I / O request.

[0051] By judging the request sent by I / 0, when the confirmation value is successful, continue to send the I / O request; when the confirmation value is failure, retry the I / O request.

[0052] Under normal conditions of the two machines: the I / O requests on the computing machine are sent to the two stores respectively through rotation training.

[0053] One storage failure: storage node failure, link failure in storage and computing.

[0054] Computing detects that the communicat...

Embodiment 3

[0056] On the basis of the above-described embodiments, the present embodiment also includes error cause judgment, and carries out error cause judgment according to the request result; when a link breaks down, then carry out a link failure mark, and repeatedly execute step 2; otherwise end the I / O of step 1 O request.

[0057] Further judge the confirmation value as a failure result, and determine whether the I / O request can be retried. By marking the link failure, the node failure can be effectively identified; at the same time, by adding a judgment and switching mechanism, under node failure , capable of high-availability switching.

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 present invention relates to the Multipath technology of Linux, and discloses a cross-node multi-link redundancy protection method and system; it includes, step 1, block device aggregation, which aggregates block devices of different links into a block device of one link , and accept the I / O request from the application; step 2, block device state judgment, judge the state of the block device aggregated in step 1 according to the I / O request in step 1, if there is a block device in active state, then the device is active; otherwise, end the I / O request in step 1. The cross-node multi-link redundancy protection method and system designed by the present invention aggregate the block devices stored and output to the computing in the computing-storage separation architecture. When one storage node is in the active state, multiple A block device can still achieve normal high-availability switching, thus ensuring the continuity of I / O and business continuity, and improving the economic benefits of the enterprise.

Description

technical field [0001] The invention relates to the Linux Multipath technology, in particular to a cross-node multi-link redundancy protection method and system in the Linux original Multipath scheme. Background technique [0002] Linux Multipath technology is based on Linux's original solution to provide block device multi-link high-availability switching and load balancing. [0003] Storage node: a standard X86 server, which is equipped with more storage media (mechanical hard disk, solid-state disk), and the computing power of the CPU is average; logically, this node provides storage capabilities, so it is collectively referred to as storage node. Computing node: a standard X86 server, on which the CPU has high computing power, and does not configure storage media itself, but uses storage mounted on other external nodes through some technical means (iscsi, NVMeOF); logically, the Nodes provide computing power, so they are collectively called computing nodes. [0004] B...

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): G06F11/20
CPCG06F11/2017
Inventor 李建辉吴炎魏兴华陈栋李春罗春臧冰凌张文件
Owner HANGZHOU WOQU NETWORK TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com