Supercharge Your Innovation With Domain-Expert AI Agents!

Quick retrieval method for variable-length data blocks

A technology of long data and data blocks, which is applied in the field of fast retrieval of variable-length data blocks, can solve problems such as taking up a large calculation time, and achieve the effect of improving retrieval efficiency and reducing the number of times

Pending Publication Date: 2021-10-12
HOHAI UNIV +2
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the fingerprint of the data block generally uses a secure hash function such as SHA256, SHA-3, etc., and calculating the fingerprint first will take up a lot of computing time

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
  • Quick retrieval method for variable-length data blocks
  • Quick retrieval method for variable-length data blocks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] Such as figure 1 As shown, a fast retrieval method for variable-length data blocks:

[0031] (1) Input a group of blocks that need repeated detection;

[0032] (2) Enter a block. If the current input is empty, terminate the search; otherwise, go to step (3)

[0033] (3) Extract the length L of the block, the first byte A, the last byte B, and the [L / 2]th byte C;

[0034] (4) Calculate K=min(L,65536)mod 256;

[0035] (5) Taking K, A, B, and C as nodes in turn to construct an index tree S-K-A-B-C, wherein S represents the root node of the index tree;

[0036] (6) If there are no other blocks under the current index, the current detection block is the only block, and return to step (2); otherwise, enter the next step;

[0037] (7) calculate the fingerprint of the data block that does not calculate the fingerprint under the current index and the currently detected data block;

[0038] (8) Compare the currently detected data block fingerprint with other block fingerpri...

Embodiment 2

[0040] Such as figure 2 As shown, a fast retrieval method for variable-length data blocks:

[0041] (1) Read a variable-length data block group to be retrieved;

[0042] (2) input a data block 1 of the variable-length data block group;

[0043] (3) Extract the length L of the data block 1 and the byte values ​​1D, 34, 50 of some positions;

[0044] (4) Map the length of data block 1 to the [0,255] interval, and calculate K=min(L,65536) mod 256, where K represents the value that maps the current data block length to the [0,255] interval, and L represents the data Block length, min(L,65536) means to take the smaller value between L and 65536, which is used to divide data blocks longer than 64KB into the same child node, and mod means to take a modulo operation;

[0045] (5) Using K, 1D, 34, and 50 as keys in turn to construct an index tree S-K-1D-34-50, wherein S represents the root node of the index tree;

[0046] (6) There are no other data blocks under the current index,...

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 quick retrieval method for a variable-length data block. The quick retrieval method comprises the following steps: extracting the length of the data block and byte values of partial positions; constructing an index tree; calculating and comparing data block fingerprints; and the like. According to the invention, the index is constructed by extracting the length and bytes of the variable-length data blocks, and the retrieval mode of firstly retrieving conflicts and then calculating fingerprints is realized, so that the fingerprint calculation process is reduced, and the retrieval efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of data storage, in particular to a fast retrieval method for variable-length data blocks. Background technique [0002] An important index to measure the deduplication capability is system overhead, which mainly includes fingerprint calculation overhead and fingerprint retrieval overhead. In the storage system, with the migration of time, the stored data will become larger and larger. At this time, fingerprint retrieval and comparison will not only occupy a large amount of computing resources, but also increase disk IO, resulting in a decrease in retrieval efficiency. Therefore, retrieval efficiency optimization is the main means to reduce system overhead. [0003] At present, retrieval efficiency optimization mainly relies on the rapid judgment of Bloom filter, pre-loading data by using data locality, constructing hierarchical index by using data similarity, optimizing index structure according to differe...

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): G06F16/2453G06F16/22G06F3/06
CPCG06F16/2453G06F16/2246G06F3/064Y02D10/00
Inventor 徐振楠吕鑫吴涛高晟凯
Owner HOHAI UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More