Distributed multi-dimensional index DAER-Tree of Append-efficient

Abstract

The invention discloses a distributed multi-dimensional index DAER-Tree of an Append-official, the index architecture of the DAER-Tree is of a Global-local type, and the distributed multi-dimensional index DAER-Tree of the Append-official comprises a Global index and a local index of the Global index; the Global index is constructed on the basis of a CAN P2P network, and the local index is an AER-Tree constructed on the basis of multi-dimensional data stored locally. According to the invention, the efficient multi-dimensional index AER-Tree is added, on the basis of the efficient multi-dimensional index AER-Tree, the C2 overlay network technology is combined, and compared with a traditional distributed multi-dimensional index technology, the index construction speed, the index size, the storage space utilization rate, the query response time, the I / O overhead and other aspects are greatly improved.

Description

technical field [0001] The invention relates to the field of data retrieval, in particular to an Append-efficient distributed multi-dimensional index DAER-Tree. Background technique [0002] In today's big data era, the data generated by many applications usually has a large amount of data, data enters the system continuously at high speed, requires large-scale (cluster) storage space, and needs to quickly index the data entering the system to meet Retrieval needs and other characteristics. Typical applications include: various surveillance applications such as video surveillance, various data stream and sensor applications, and applications that need to analyze a large number of generated (event) logs. In these application scenarios, the data that needs to be archived into the system may continue to be retrieved and analyzed in the future. In order to improve data access performance, corresponding (multidimensional) indexes are usually established when archiving these data...

AI Technical Summary

Problems solved by technology

There are two problems with this type of technology: one is that it may violate the data management strategy of the enterprise

However, when the load of the system increases to a certain extent and the load of a single node in the distributed system is still very heavy, if the distributed index still uses the existing B + -Tree and R-Tree technologies, the overhead of splitting index nodes generated when continuously inserting data will still be large, and will affect the performance of data insertion and index construction

Images