DocumentCode :
2748272
Title :
A network status predictor to support dynamic scheduling in network-based computing systems
Author :
Kim, JunSeong ; Lilja, David J.
Author_Institution :
Electr. & Comput. Eng., Minnesota Univ., Minneapolis, MN, USA
fYear :
1999
fDate :
12-16 Apr 1999
Firstpage :
372
Lastpage :
378
Abstract :
The management of networks has often been ignored in network-based computing systems due to the difficulty of estimating application programs´ network latency and bandwidth requirements, and the difficulty of predicting the system network load. To help address this deficiency and thereby support dynamic network resource scheduling, we propose the network status predictor (NSP). This tool is a general and extensible network load monitor that introduces lower and upper latency prediction bounds. We evaluate its ability to dynamically predict TCP/IP end-to-end latency, with varying network loads using a cluster of SGI multiprocessors interconnected with a fibre channel network. Our results show that a combination of numerical predictors can be dynamically selected based on the network´s recent state to produce better predictions than when using a single predictor alone
Keywords :
local area networks; processor scheduling; transport protocols; TCP/IP end-to-end latency; bandwidth requirements; cluster of SGI multiprocessors; dynamic network resource scheduling; dynamic scheduling; fibre channel network; network latency; network load monitor; network status predictor; network-based computing systems; numerical predictors; system network load; Accuracy; Computer networks; Delay; Dynamic scheduling; Engineering management; History; Intelligent networks; Monitoring; Processor scheduling; Resource management;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Parallel Processing, 1999. 13th International and 10th Symposium on Parallel and Distributed Processing, 1999. 1999 IPPS/SPDP. Proceedings
Conference_Location :
San Juan
Print_ISBN :
0-7695-0143-5
Type :
conf
DOI :
10.1109/IPPS.1999.760503
Filename :
760503
Link To Document :
بازگشت