Method for active database defence for multi-GPU parallel processing

An active defense, database technology, applied in the direction of digital data processing, multi-programming device, resource allocation, etc., can solve problems such as difficult to meet data processing, and achieve the effect of reducing data transmission overhead and ensuring correctness

Active Publication Date: 2017-08-29
深圳市知穹科技有限公司
View PDF2 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing database active defense system implemented by CPU thread technology has been difficult to meet the needs of data processing in the face of high concurrent network traffic

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
  • Method for active database defence for multi-GPU parallel processing
  • Method for active database defence for multi-GPU parallel processing
  • Method for active database defence for multi-GPU parallel processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The invention will be described in detail below in conjunction with the accompanying drawings and embodiments.

[0025] Due to its super computing power, high-speed memory access bandwidth, and high data-level parallel architecture, modern graphics processing units (GPUs) are widely used in computing environments as high-performance computing components, and modern computing technology is gradually entering the The era of data-level parallel computing.

[0026] Whether data-level parallel computing can be more and more widely used in practical applications depends not only on whether the designed and implemented data-level parallel algorithms can obtain high computing power in the current hardware computing environment, but also on the basis of And what is important is whether the data-level parallel algorithm designed and implemented can have good and sustainable scalability at the system level (between nodes) and node level (inside a single node), that is, its perform...

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 discloses a method for active database defence for multi-GPU parallel processing. The method is characterized in that video memories of all the GPUs are virtualized into a shared video memory and established into a global array of a GPU side; a shared cache area is set in CPU; when a network data packet in the shared cache area reaches a judgment threshold for starting GPU processing, a data packet of the shared cache area will be synchronized to the global array of the GPU side; and then, the CPU will dispatch kernel processing functions of the multiple GPUs to process the network data packet. According to the invention, multi-GPU acceleration processing is adopted when the database active defence system faces a high parallel network flow rate, so that overall performance can be improved, and problems such as high rate and low delay faced by the CPU at the moment can be solved.

Description

technical field [0001] The invention relates to a network data packet parallel processing technology for accessing a database, in particular to a database active defense method for multi-GPU parallel processing. Background technique [0002] With the rapid development of Internet network speed, people put forward higher requirements for high-speed and low-latency network access. On the one hand, due to the subsequent introduction of multi-thread communication overhead and synchronization overhead on the CPU, the performance improvement is small. There is even a phenomenon of not rising but falling; on the other hand, the cost of improving CPU processing performance is too high. The existing database active defense system implemented by CPU thread technology has been difficult to meet the data processing requirements in the face of high concurrent network traffic. Contents of the invention [0003] The purpose of the present invention is to provide a multi-GPU parallel pro...

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): G06F9/50
CPCG06F9/5027
Inventor 秦明尹立东谢东辉
Owner 深圳市知穹科技有限公司
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