DMCA
Scheduling for Cache Affinity in Parallelized Communication Protocols (1994)
Cached
Download Links
- [ftp.cs.umass.edu]
- [gaia.cs.umass.edu]
- [gaia.cs.umass.edu]
- DBLP
Other Repositories/Bibliography
| Venue: | In Proceedings of 1995 SIGMETRICS/Performance International Conference on Measurement and Modeling of Computer Systems |
| Citations: | 8 - 4 self |
BibTeX
@TECHREPORT{Salehi94schedulingfor,
author = {James D. Salehi and James F. Kurose and Don Towsley},
title = {Scheduling for Cache Affinity in Parallelized Communication Protocols},
institution = {In Proceedings of 1995 SIGMETRICS/Performance International Conference on Measurement and Modeling of Computer Systems},
year = {1994}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
In this paper, we explore the benefits of processor-cache affinity scheduling of parallel network protocol processing, in a setting in which protocol processing executes on the multiprocessor host concurrently with a general workload of non-protocol activity. We implement two parallelization approaches, Locking and Independent Protocol Stacks (IPS), in the UDP/IP/FDDI protocol stack in the x-kernel on an SGI Challenge XL multiprocessor. We then conduct a set of multiprocessor experiments designed to measure packet processing times under specific conditions of cache state. These measurements are then used to parameterize the analytic component of a simulation model of multiprocessor protocol processing, designed to reflect the impact on protocol execution time of the non-protocol workload. In this manner, we are able to explore the benefits of affinity scheduling of parallel network processing (while accounting for the impact of general non-protocol activity) in a carefully-controlled e...
Keyphrases
parallelized communication protocol cache affinity udp ip fddi protocol stack independent protocol stack sgi challenge xl multiprocessor general non-protocol activity multiprocessor protocol processing protocol execution time multiprocessor host parallelization approach general workload processor-cache affinity scheduling parallel network protocol processing packet processing time analytic component cache state multiprocessor experiment specific condition affinity scheduling non-protocol activity protocol processing simulation model non-protocol workload parallel network processing
