期刊名称:International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN:2320-9798
电子版ISSN:2320-9801
出版年度:2017
卷号:5
期号:5
页码:9338
DOI:10.15680/IJIRCCE.2017.0505111
出版社:S&S Publications
摘要:Accurately estimating congestion for proper global adaptive routing decisions (i.e., determine whether apacket should be routed minimally or non-minimally) has a significant impact on overall performance for high-radixtopologies, such as the Dragonfly topology. Prior work have focused on understanding near-end congestion i.e.,congestion that occurs at the current router – or downstream congestion – i.e., congestion that occurs in downstreamrouters. However, most prior work do not evaluate the impact of far-end congestion or the congestion from the highchannel latency between the routers. In this work, we refer to far-end congestion as phantom congestion as thecongestion is not “real” congestion. Because of the long inter-router latency, the in-flight packets (and credits) result ininaccurate congestion information and can lead to inaccurate adaptive routing decisions. In addition, we show howtransient congestion occurs as the occupancy of network queues fluctuate due to random traffic variation, even insteady-state conditions. This also results in inaccurate adaptive routing decisions that degrade network performancewith lower throughput and higher latency. To overcome these limitations, we propose a history-window basedapproach to remove the impact of phantom congestion. We also show how using the average of local queueoccupancies and adding an offset significantly remove the impact of transient congestion. Our evaluations of theadaptive routing in a large-scale Dragonfly network show that the combination of these techniques results in anadaptive routing that nearly matches the performance of an ideal adaptive routing algorithm.