Novel concurrent memory data organization and access method

Abstract

The invention relates to the field of data processing technology, in particular to a novel concurrent memory data organization and access method. The method is characterized in that the memory data organizational form provides two types of data access methods: Key-Value accurate index method and Key-ValueList clustered index method; any attribute of each entity can be accessed according to requirements; a multidimensional index is simultaneously supported and a single entity can be retrieved through different indexes; the entire memory data can be managed in a manner of block organization; management information of different levels and different types is stored respectively in each memory block; and data access is controlled in series based on the logic of each memory block. Compared with the prior art, the method has the advantages that the memory data structure is flexibly designed and organized on the premises of data integrity and high concurrent access efficiency, an efficient data retrieval model is provided, the WORM (write-once-read-many) concurrent data access is supported, the integrity of data read and write transaction is effectively ensured, the resource consumption is effectively reduced, and the concurrent access ability is improved.

Description

[technical field] [0001] The invention relates to the technical field of data processing, in particular to a novel concurrent memory data organization and access method, which is mainly applied to memory data access scenarios requiring certain efficiency in data organization and access. [Background technique] [0002] Data storage and access are the most common needs in various computer applications. With the continuous improvement of memory capacity, more and more data is loaded into the memory. The system adjusts and sets the data status by reading and writing the memory. , how to organize the memory data so that it can meet the requirements of efficient concurrent access, and can still ensure the transactional consistency of data reading and writing during the concurrent access of data. Being able to meet the data access and positioning requirements of different application scenarios has become one of the common problems faced by most application systems. [0003] Taking...

AI Technical Summary

Problems solved by technology

[0015] However, the existing technical methods do not combine the two types of index methods to provide complete support. Some methods only provide a single non-unique index retrieval method, and do not support the actual multi-dimensional index; some methods provide multi-dimensional index, but in the multi-dimensional index There are deficiencies in data access and operation. Some methods do not fully support the addition, deletion and update of multidimensional indexes, but only support the addition and update functions; some methods only consider the construction and deletion logic of various indexes under normal circumstances. , did not fully consider the reasonable processing method in the case of the unexpected termination of the modification operation during the data access process, resulting in local inconsistency of the memory data; some methods, while providing multidimensional indexes, did not solve the problem without reducing the concurrency performance On the issue of ensuring the integrity of read and write transactions, in short, there are defects

Images