Flashback method and device of distributed object-based storage system

An object storage and distributed technology, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve problems such as high complexity, difficult to achieve flashback speed, and high energy consumption.

Active Publication Date: 2017-03-08
FOCUS TECH
View PDF8 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

There are several problems with this approach: First, it takes a lot of energy to study the underlying implementation details of a certain storage engine. When the underlying implementation of the storage engine changes, it is necessary to re-evaluate the impact of these changes on the specific implementation of the flashback feature, and may even need to re-implement the flashback feature; finally, such a mining method for a certain storage engine, Does not have a certain degree of universality, and the cost performance is not high
The methods proposed in the above-mentioned patents solve the problem of data backup and rollback to a certain extent, but they all need to establish an additional data mirror server, an...

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
  • Flashback method and device of distributed object-based storage system
  • Flashback method and device of distributed object-based storage system
  • Flashback method and device of distributed object-based storage system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] Example 1, write the data whose business primary key is 100 for the first time, the specific steps are as follows:

[0070] Step 202: After receiving the data write request, the storage system calls Version-generator 101 to generate a unique version number v1;

[0071] Step 203: the storage system writes the object data ontology to the Primary-database 102, and the storage primary key is designed to be 100~v1~t1. The result is like image 3 Shown in 301.

[0072] Step 204: Since the record with the business primary key 100 is written for the first time, the latest record of the data with the business primary key 100 is 100~v1~t1 according to the calculation of the Primary-database 102.

[0073] Steps 205-206: Record 100-v1-t1 in the Servering-database 104, use 100-v1-t1 as the primary key for recording, and write null values ​​for the corresponding data. The result is like image 3 Shown in 303.

[0074] Step 207: Since there is no historical version for the record...

Embodiment 2

[0075] Embodiment 2, updating the data whose business primary key is 100, the specific steps include:

[0076] Step 202: Since the data whose business primary key is 100 has been written before, when the storage system receives the write request, it calls Version-generator 101 to generate the unique version number of the data as v2;

[0077] Step 203: the storage system writes the object data ontology to the Primary-database 102, and the primary key is designed to be 100~v2~t2. The result is like Figure 3-3 04.

[0078] Step 204: Since v2 is the latest version with the business primary key of 100, the latest data records of the data with the business primary key of 100 are 100~v2~t2 through the calculation of the Primary-database 102.

[0079] Steps 205-206: Record 100-v2-t2 in the Servering-database 104. The result is like Figure 3-3 06.

[0080] Step 207: Compared with the data stored with the primary key of "100-v2-t2", the stored primary key of "100-v1-t1" is the hi...

Embodiment 3

[0081] Example 3: Delete the data whose business primary key is 100, and the data version is v2. The specific steps are as follows:

[0082] Step 210: Receive the record request for deleting the business primary key of 100 sent by the system, find the service version with the business primary key of 100 in the Servering-database 104, and delete the version (v2) according to the request, from the Servering-database 104 Remove 100~v2~t2;. The result is like image 3 Shown in 309.

[0083] Step 211: Write the record with 100-v2-t2 as the primary key in the Back-database 103. The result is like Figure 3-3 08. However, the data in Primary-database 102 remains unchanged during this delete operation, and the result is as follows Figure 3-3 07.

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

A flashback method of a distributed object-based storage system is characterized by comprising the following steps: performing writing and reading of data of the distributed object-based storage system, data deleting data flashing bac, and deleting of outdated version data; constructing three data storing units, namely, a primary database, a back database and a servering database, based on a storage engine of a key-value data model; and performing data writing and reading, data deleting, data flashing back and deleting of outdated version data based on the three data storage units when a storage system receives a data operation request. With the adoption of the flashback method, the system complexity, increased by the flashback property, of the data can be greatly reduced, and a user can avoid the problems of additional data mirror server, serializing and deserializing of mirror files; a record-level previous version recovery mode is treated as the flashback mechanism, so that the data flashback process is only performed on a specified data set, and as a result, mis-operation can be avoided.

Description

technical field [0001] The invention belongs to the technical field of massive data processing, and in particular relates to a data flashback method and device of a distributed object storage system, which solves the problems of version flashback and expired version cleaning of massive object data. Background technique [0002] In recent years, the rise of new applications such as blogs, wikis, and spaces has led to changes in the way Internet content is provided, giving birth to the era of web 2.0 in which users create content, and driving video applications , online games, search engines and other Internet-derived businesses have developed rapidly. These applications generate a large amount of object data every day, including pictures, documents, videos, etc. As these object data play an increasingly important role in the application (it is hard to imagine what a platform like Taobao will be like without pictures ), in many scenarios, we increasingly hope that the storage...

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
IPC IPC(8): G06F17/30
CPCG06F16/215G06F16/219
Inventor 梁峰
Owner FOCUS TECH
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