Method, device and system for network package processing

A network packet and graphics processor technology, applied in the computer field, can solve the problems of low network packet processing efficiency, high resource consumption, long response time, etc., and achieve the effect of reducing the number of copies, saving system resources, and improving efficiency.

Active Publication Date: 2013-07-10
国久大数据股份有限公司
View PDF3 Cites 37 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Although this solution can accelerate network protocol tasks in the kernel, such as routing lookup, SSL (Secure Sockets Layer, secure socket layer), IPSec (Internet Protocol Security, network protocol security), DPI (Deep Packet Inspection, deep

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, device and system for network package processing
  • Method, device and system for network package processing
  • Method, device and system for network package processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] An embodiment of the present invention provides a method for network packet processing, specifically as figure 1 As shown, the method includes:

[0069] 101. Receive a network packet, and cache it in a first address of a memory buffer.

[0070] It should be noted that the "first" in the first address does not have any special meaning, and only indicates that it is a specific address in the memory buffer.

[0071] 102. According to the first address of the memory buffer and the address mapping relationship between the address of the pre-stored memory buffer and the physical address in the video memory of the graphics processor GPU, determine the address of the network packet in the video memory of the GPU physical address.

[0072] Specifically, the embodiment of the present invention is based on the unified virtual address space UVA (Unified virtual addressing, unified virtual address space) technology, the memory resources and the video memory resources on the graphi...

Embodiment 2

[0106] An embodiment of the present invention provides a method for network packet processing, specifically as image 3 The illustrated network packet processing system 300 is described below. The network packet processing system may specifically include a network card 301, a graphics card 302, a CPU 303, a memory 304, a chipset 305, a PCIe bus 306, and a DMA controller 307, wherein the The graphics card 302 includes a GPU3021 and a video memory 3022. The network card 301 and the graphics card 302 are connected to the same PCIe bus 306 and communicate with the CPU 303 through the chipset 305 . The method is as Figure 4 shown, including:

[0107] 401. The network card receives the network packet and caches it in the first address of the memory buffer.

[0108] 402. The network card determines that the network packet is in the video memory of the GPU according to the first address of the memory buffer and the address mapping relationship between the address of the pre-stored...

Embodiment 3

[0190] An embodiment of the present invention provides an apparatus 700 for network packet processing, specifically as Figure 7 As shown, it includes a receiving unit 701 , a storage unit 702 , a determining unit 703 , a control unit 704 , and a sending unit 705 .

[0191] The receiving unit 701 is configured to receive network packets.

[0192] The storage unit 702 is configured to cache the network packet in the first address of the memory buffer.

[0193] The determining unit 703 is configured to determine, according to the address mapping relationship between the first address of the memory buffer and the address of the pre-stored memory buffer and the physical address in the video memory of the graphics processor GPU, where the network packet is located. Specifies the physical address in the video memory of the GPU.

[0194] The control unit 704 is configured to control a direct memory access DMA controller, and write the network packet into the determined physical add...

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

An embodiment of the invention provides a method, a device and a system for network package processing, response time of the system can be reduced, system resources are saved, and the efficiency of the network package processing is improved. The method includes the steps that a network package is received and is cached into a first address of an internal memory buffer area; according to the address mapping relation among the first address of the internal memory buffer area, an address of a prestored internal memory buffer area and a physical address of a video memory of a graphics processor (GPU), the physical address of the network package in the video memory of the GPU is determined; a direct memory access (DMA) controller is controlled to write the network package into the determined physical address of the video memory of the GPU where the network package is in; and first indication information is sent to enable a communication opposite end to process the network package according to the first indication information, wherein the first indication information carries the determined physical address of the network package in the video memory of the GPU. The method, the device and the system are applied to the field of computers.

Description

technical field [0001] The invention relates to the field of computers, in particular to a method, device and system for processing network packets. Background technique [0002] With the diversification of network services and applications, the methods of network packet processing are also diversified. The traditional method of using a CPU (Central Processing Unit, central processing unit) to process network packets can no longer meet the requirements of network transmission speed. [0003] At the same time, GPU (Graphics Processing Unit, Graphics Processing Unit) is developing at a speed exceeding Moore's Law, and its floating-point computing capability far exceeds that of CPU. The aspect has been greatly improved, and it has the characteristics of stream processing, high-density parallel computing, and programmable pipeline. The platform for general computing using GPU at the software level is also becoming more and more mature. In particular, the open unified standard O...

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
IPC IPC(8): H04L12/861H04L29/08H04L29/12
Inventor 王继海
Owner 国久大数据股份有限公司
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