A reading method and device in a distributed system

A distributed system and reading device technology, applied in the direction of file system, instrument, file access structure, etc., can solve the problems of increased queuing, failure to optimize glitch rate, and increased glitch rate

Active Publication Date: 2020-11-20
ALIBABA GRP HLDG LTD
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0018] The sending interval in the scene with low pressure is too long, because the redundant request is not sent in time, so a better glitch rate optimization effect cannot be obtained;
[0019] The sending interval is too small in stressful scenarios, because the sending of redundant requests also consumes resources. In stressful scenarios, this kind of consumption will not only fail to optimize the glitch rate, but will increase more queues and waits, resulting in an increase in the glitch rate A vicious circle with increased sending of redundant requests reduces system stability

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 reading method and device in a distributed system
  • A reading method and device in a distributed system
  • A reading method and device in a distributed system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] Embodiment 1. A reading method in a distributed system, such as figure 2 As shown, including steps S110-S120:

[0070] S110. After receiving the user's read request for the first file, send a read data request to one of the storage nodes corresponding to the first file;

[0071] S120. When the waiting time for a response reaches or exceeds a first time interval, send a redundancy request for the read data request to another storage node corresponding to the first file;

[0072] Wherein, the first time interval is dynamically determined according to the performance index of the read operation.

[0073] In this embodiment, the sending interval of the redundancy request for the read data request (that is, the first time interval) is no longer fixed, but dynamically adjusted according to the situation. Dynamic adjustment means that the first time interval will follow the read The performance index of the operation changes; since the performance index of the read operatio...

other Embodiment approach

[0098] In other implementation manners, another method can also be used to realize the dynamic determination of the first time interval, such as using the delay of the read operation and sending the redundancy request within a predetermined length of time before the moment of the first time interval to be determined. The accumulated value of the number of times to calculate the first time interval; for example, according to the number of redundant requests sent and the number of read operations within a predetermined length of time before the time of the first time interval to be determined, the The average number of sending times of the read operation is then used together with the average value of the delay to calculate the first time interval.

[0099] In this embodiment, the first time interval is based on L avg And the Qps calculation can include:

[0100] The first time interval is equal to the function value of the predetermined function; the argument of the predetermi...

Embodiment 2

[0125] Embodiment 2. A reading device in a distributed system, such as image 3 shown, including:

[0126] The first request module 21 is configured to send a read data request to one of the storage nodes corresponding to the first file after receiving the user's read request for the first file;

[0127] The second request module 22 is configured to send a redundancy request for the read data request to another storage node corresponding to the first file when the waiting time for a response reaches or exceeds a first time interval; wherein, the The first time interval is dynamically determined according to the performance index of the read operation.

[0128]In this embodiment, the first request module 21 is the part responsible for initiating a read data request in the above device, which may be software, hardware or a combination of both.

[0129] In this embodiment, the second request module 22 is a part in the above device responsible for initiating a redundancy request...

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 discloses a reading method and apparatus for a distributed system. The reading method comprises the steps of sending a data reading request to one of storage nodes corresponding to a first file after the reading request of a user to the first file is received; and when the length of time of waiting for a response reaches or exceeds a first time interval, sending a redundancy request for the data reading request to another storage node corresponding to the first file, wherein the first time interval is dynamically determined according to a performance index of reading operation. According to the method and the apparatus, a sending interval of the redundancy request for the data reading request can be adaptively adjusted.

Description

technical field [0001] The invention relates to the field, in particular to a reading method and device in a distributed system. Background technique [0002] Aiming at the optimization of read glitches in distributed systems, a strategy of BackupRequest is proposed in the related art. BackupRequest is a redundant request added in the ordinary read process. It reduces the user-perceived read delay glitch rate on the basis of not requiring a reduction in the read delay of a single request. [0003] Take Google's Google File System (Google File System, GFS) as an example to describe the reading process including BackupRequest. The specific structure of Google File System is as follows: figure 1 Shown: [0004] Client library: provides various interfaces for users of distributed storage systems; [0005] Chunkserver: As a data management module, it specifically manages user data; [0006] Master: As a namespace management module, it manages the meta data of the distributed ...

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): G06F16/182G06F16/13
CPCG06F16/13G06F16/182
Inventor 刘善阳
Owner ALIBABA GRP HLDG LTD
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