AVL tree-based data writing method and apparatus

Abstract

The present invention discloses an AVL tree-based data writing method and apparatus, which use a read-write lock for performing concurrency control on data writing. The method comprises: receiving a data write request; determining a search path according to sizes of written node value and characteristics of an AVL tree; determining whether each node in the search path is in a lock state, wherein the lock state is that when a first node in a currently concurrent write operation is a root node of a subtree locally adjusted, the first node is set into a locked state; If no node in the search path is in the lock state, performing the write operation; and if a certain node in the search path is in the lock state, blocking the write operation up to the locked node until the lock state of the locked node is canceled. According to the AVL tree-based data writing method and apparatus provided by the present invention, parallel processing is allowed to be performed on the write operation without a conflict in the local of the AVL tree, thereby improving data writing performance.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to an AVL tree-based data writing method and device. Background technique [0002] AVL tree, also known as Balanced Binary Tree, has the following properties: it is an empty tree or the absolute value of the height difference between its left and right subtrees does not exceed 1, and both left and right subtrees are a balanced binary tree. Because of its special structure, it has better performance than linear data structures in indexing and storing single-threaded data; however, the lock mechanism introduced by concurrency protection is not perfect during multi-threaded access, resulting in performance bottlenecks. [0003] Red-black tree (RedBlackTree) is a self-balancing binary search tree. Similar to AVL tree, the red-black tree maintains the balance of the binary search tree through specific operations during insertion and deletion operations, thereby obtaining a higher searc...

AI Technical Summary

Problems solved by technology

However, this processing method does not conflict with the search path of the two requests to modify the node, that is to say, the request that the search path does not conflict can be issued at the same time, but it must wait for the previous request to be completed before it can be issued. , the time consumed by this waiting process brings additional performance overhead, and it may not be two requests, but more requests, then the performance overhead will be even greater

Images