Adaptive cardinal number tree dynamic indexing method based on GPU parallelism

A dynamic index and radix tree technology, applied in other database indexes, other database retrieval, digital data information retrieval, etc., can solve the problems of inability to change node types, inability to operate in parallel, and inability to make full use of warps and caches.

Active Publication Date: 2020-11-27
NORTHEASTERN UNIV
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The search, insertion and deletion operations on the four types of nodes of the adaptive radix tree index are all single-threaded serial designs, which cannot be operated in parallel, and the existing GPU method is a serial tree structure, which cannot dynamically update nodes. The type changes, and the warp and cache cannot be fully utilized when querying, and the performance is poor

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Adaptive cardinal number tree dynamic indexing method based on GPU parallelism
  • Adaptive cardinal number tree dynamic indexing method based on GPU parallelism
  • Adaptive cardinal number tree dynamic indexing method based on GPU parallelism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] The invention will be further described below in conjunction with the accompanying drawings and specific implementation examples.

[0047] A GPU-parallel adaptive radix tree dynamic indexing method is a GPU-parallel-based dynamic index design that makes full use of the parallel computing capabilities of the GPU and combines the adaptive radix tree structure to build, insert, delete, and Query operations are parallelized, and the indexing method supports high-throughput, low-latency batch parallel update and query operations. Firstly, according to the characteristics of data distribution in the radix tree, a batch of data to be processed is firstly sorted by high-order radix first, and then multiple grouping operations are performed after the sequence is in order. For the 32-bit key, the highest 8-bit is counted for the first time. , divide each data sequence to be processed into a maximum of 256 segments, so that each segment has the same prefix inside, and then process...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides an adaptive cardinal number tree dynamic indexing method based on GPU parallelism. Firstly, an adaptive cardinal number tree data structure is constructed; the first two layerscreate tree nodes of the Node256 type; a cardinal number sequencing method based on high-order priority is created for the third layer and the fourth layer; tree nodes capable of accommodating corresponding sizes are created according to the number of the branches; the creation of a dynamic data structure is realized; it can be ensured that the latest update in the original batch of data is stillbehind the old update after sorting; and de-duplication is performed, redundant old updates are removed, latest updates are reserved, each section of sequence without duplicate data is inserted into anode corresponding to the section after duplicate removal to finish the establishment of the whole self-adaptive cardinal number tree, and secondly, data insertion, query and deletion operations canbe performed in parallel based on GPU parallel computing capability.

Description

technical field [0001] The invention relates to the technical field of parallel indexing in the direction of computer databases, in particular to a GPU parallel adaptive radix tree dynamic indexing method. Background technique [0002] Indexing technology is one of the key technologies in modern information retrieval, search applications, and data mining. In order to accurately locate data in a large amount of data, various types of data indexes have been proposed for different types of data and different query requirements. In recent years, the explosive growth of data scale has further expanded the field of research to large-scale, high-dimensional, and sparse data sets. Efficient index construction and data query for such data sets have become the main research direction at this stage. . In addition, with the expansion of the data scale, the requirements for the processor are getting higher and higher. The parallel capability of the graphics processing unit (GPU) provide...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/901G06F16/245
CPCG06F16/24569G06F16/9027
Inventor 谷峪宛长义李传文宋振于戈
Owner NORTHEASTERN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap