A multi-level disordered indexing method and system

Abstract

The invention discloses a rapid multi-level disordered indexing method and a rapid multi-level disordered indexing system. In a key value system constructed based on a storage device, all keys are organized into a B + tree, the keys are quickly indexed by adopting a sequential method or an improved dichotomy, and specific keys and value addresses thereof in leaf nodes are searched, so that orderedindexing is realized through multi-level disordered leaf nodes. The system comprises a first storage device and a second storage device, which are used for storing the non-leaf nodes and the leaf nodes of B + trees respectively; The same or different storage devices can be adopted respectively; Or mutually isolated storage areas are divided in the same storage device. According to the invention,through multistage disordered indexing, ordered indexing is ingeniously realized, so that the method has the advantages of ordered indexing and disordered indexing, and meanwhile, the requirements andeffects of reading, writing and deleting operations on high performance are met.

Description

technical field [0001] The invention relates to the fields of big data processing technology and storage index technology, in particular to a fast multi-level disordered index method and system. Background technique [0002] With the advent of the big data era, the Key-Value storage system (KV) ushered in rapid development. The key-value system assigns a globally unique key (Key) to each value (Value) when performing a write operation for the first time. In subsequent read and delete operations, the key needs to be obtained first. Therefore, the fast indexing speed of the key directly affects or even determines the performance of the key-value system. [0003] The existing fast index methods mainly include Hash index and B+ tree index: Hash index does not support range search and has hash conflicts, so the application range is narrow. B+ tree index supports range search, and there is no problem of hash collision, so it has become a mainstream method at present. [0004] ...

AI Technical Summary

Problems solved by technology

[0005] However, the two existing technologies for leaf node organization have serious disadvantages, and cannot meet the high-performance requirements of read, write, and delete operations at the same time

One is the ordered index, the principle of which is the simplest, the most well-known, and the most widely used, and has become common knowledge in this field, such as figure 1 As shown, its advantage is that the performance of read operation is high, and the disadvantage is that the performance of write and delete operation is low; the second is an unordered index, such as CN201310456838.7, such as figure 2 As shown, the advantage is high performance of write and delete operations, and the disadvantage is low performance of read operations

Images