Pipelined packet processing

Details Pipelined packet processing Pipelined packet processing

2002-02-25

2004-12-28

El-Hady, Nabil (Department: 2154)

Electrical computers and digital data processing systems: input/

Input/output data processing

Concurrent input/output processing and data transfer

C710S021000, C710S022000, C709S232000, C709S248000, C709S241000

Reexamination Certificate

active

06836808

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Technical Field
The present invention relates in general to processing of packetized data, and in particular, to a method and system for pipelining particular packet processing functions within a packet processing node.
2. Description of the Related Art
A Fibre Channel is a data transfer architecture developed by a consortium of computer and mass storage device manufacturers and is currently a standard promulgated by the American National Standards Institute (ANSI). Fibre Channel is a high-speed networking technology used to build storage area networks (SANs). Although Fibre Channel can be utilized as a general-purpose network carrying Asynchronous Transfer Mode (ATM), Internet Protocol (IP), and other protocols, it has been primarily used for transporting Small Computer System Interface (SCSI) traffic from servers to disk arrays.
Fibre Channel is designed for new mass storage devices and other peripheral devices that require very high bandwidth. Using optical fiber to connect devices, Fibre Channel supports full-duplex data transfer rates of 2 GB/s. Fibre Channel transmission is performed serially over fiber optic links, and may eventually replace SCSI for high-performance storage systems.
There are two primary sources of latency on any packet-based network such as Fibre Channel networks: link propagation latency due to link length and node latency. The goal of the present invention is to reduce node latency with a corresponding increase in Fibre Channel throughput. To increase end-to-end Fibre Channel throughput, there exists a need to provide a method and system that pipeline protocol handler tasks. The present invention addresses such a need by providing a packet processing system applicable within packet protocol handlers that efficiently pipelines packet processing tasks and queueing algorithms to improve overall network transmission efficiency.
BRIEF SUMMARY OF THE INVENTION
A method and system for increasing the efficiency of packet processing within a packet protocol handler are disclosed herein. In accordance with the method of the present invention packet processing tasks are performed on multiple processors or threads concurrently and in a pipelined fashion. Subsequent protocol packet processing tasks for processing a single packet are performed on multiple processors or threads, acting as stages of a pipeline. The assignment of tasks to processors or threads is performed dynamically, by checking the availability of a processor or thread in the subsequent pipeline stage. The availability determination includes determining the available capacity of the input work queue associated with each processor or thread. If the subsequent pipeline stage is overloaded, the task is assigned to another processor or thread that is not overloaded. Such dynamic allocation of processors or threads to tasks within an adjustable processing pipeline maximizes efficient processing resource allocation and reduces pipeline imbalance, ensuring a flexible solution and maximal throughput in packet processing.
All objects, features, and advantages of the present invention will become apparent in the following detailed written description.


REFERENCES:
patent: 4837676 (1989-06-01), Rosman
patent: 5235595 (1993-08-01), O'Dowd
patent: 5634015 (1997-05-01), Chang et al.
patent: 5860019 (1999-01-01), Yumoto
patent: 5938736 (1999-08-01), Muller et al.
patent: 6147996 (2000-11-01), Laor et al.
patent: 6173333 (2001-01-01), Jolitz et al.
patent: 6178488 (2001-01-01), Manning
patent: 6665755 (2003-12-01), Modelski et al.
patent: 2002/0054594 (2002-05-01), Hoof et al.
patent: 2002/0120828 (2002-08-01), Modelski et al.
patent: 2003/0041163 (2003-02-01), Rhoades et al.

Affiliated with

Also associated with

Rating

Pipelined packet processing does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Pipelined packet processing, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Pipelined packet processing will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3285952