An index query method and system for a key-value storage system

Abstract

The present invention discloses an index query method and system oriented to a Key-Value storage system. The method includes: S1, obtaining Key-Value data items and storing them in a block, and generating a prefix complete binary tree according to the block nodes; S2, obtaining the key value, according to The height of the binary tree determines the key value using the digit k, and the key value is binary; S3. Obtain the first k digits of the key value, traverse from the first k digits of the key value, retrieve the first "1" in the first k digits of the key value, and record the first The digit j before a "1"; S4, judge whether j is less than k, if j is less than k, then the target result i=2 j , execute S5; if j is not less than k, then the target result i=2 k‑1 ‑1, execute S6; S5, search circularly from the j+2th bit of the key value, when "1" is retrieved, i=2×(i+1); otherwise, i=2×i+1; until j=k , stop searching to get the target result i, execute S6; S6, search block[i] in the prefix complete binary tree through the value of i, judge whether block[i] is not empty, if block[i] is not empty, output the target result i value; if block[i] is empty, execute S3 after setting k=k-1 until k=0.

Description

technical field [0001] The invention relates to the technical field of data index query, in particular to an index query method and system oriented to a Key-Value storage system. Background technique [0002] With the rapid development of the Internet, the world today has entered the era of "big data". The latest report of the international data company IDC predicts that within ten years from 2014, the total amount of data generated globally every year will increase by 40%, or about The total amount of data will double every two years, and by 2020 the total amount of global data will reach 44ZB. With the rapid growth of data volume, the form of data is also developing from traditional structured data to unstructured or semi-structured data forms. Traditional relational databases are stretched for the storage capacity of these data. Indexing, web serving of high-traffic sites, and sending streaming media data. In addition, for a specific system, most of the retrieval is bas...

AI Technical Summary

Problems solved by technology

[0003] Key-value databases generally use a tree structure to organize data. With the advent of the era of big data, it is necessary to find specific data in massive amounts of data. If the traditional tree traversal form is used, it will undoubtedly cause huge time overhead. , so an algorithm that can quickly index specific data is essential

Images