By Topic

A trace driven study of the effectiveness of cache mechanism for network processors

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

5 Author(s)
Zhen Liu ; Dept. of Comput. Sci. & Eng., Tsinghua Univ., Beijing, China ; Hao Che ; Kai Zheng ; Shanzhen Chen
more authors

Network processor (NP) is a new breed of packet forwarding engine that is designed to meet the simultaneous demands of high speed and great flexibility of today's network equipments. Cache mechanism has attracted more and more attentions in the field of NP design because of its big success in terms of latency hiding. In this paper, we evaluate the effectiveness of cache mechanism for NP under the new context of packet forwarding using real-life packet traces. Our results suggest that caching can reduce the packet loss rate to an extremely low level under reasonable hit rate and packet buffer size. However, the possibility of incurring large queuing delay makes it unsuitable for applications demanding stringent quality of service (QoS) enforcement. Furthermore, we also show that the combination of cache mechanism with multithreading will be a promising way because of the dramatically decreased both queuing and processing delay, in addition to the even lower packet loss rate.

Published in:

Wireless Communications, Networking and Mobile Computing, 2005. Proceedings. 2005 International Conference on  (Volume:2 )

Date of Conference:

23-26 Sept. 2005