TY - JOUR
T1 - A Performance Comparison of Coscheduling Strategies for Workstation Clusters
AU - ANGLANO, Cosimo Filomeno
PY - 2001/1/1
Y1 - 2001/1/1
N2 - Workstation clusters are emerging as a general-purpose computing platform for the execution of workloads comprising parallel and sequential applications. The scalability and flexibility typical of implicit coscheduling strategies makes them a very promising solution to the scheduling needs of workstation clusters. In this paper we present a simulation study that compares, for a variety of workloads (that include both parallel and sequential applications) and operating system schedulers, 12 implicit coscheduling strategies in terms of the performance they are able to deliver to applications. By using a detailed simulator, we evaluate the performance of different coscheduling alternatives for a variety of simulation scenarios, and we identify the set of strategies that deliver the best performance to all the applications composing typical cluster workloads. Moreover, we show that for schedulers providing i>immediate preemption, the best strategies are also the simplest ones to implement.
AB - Workstation clusters are emerging as a general-purpose computing platform for the execution of workloads comprising parallel and sequential applications. The scalability and flexibility typical of implicit coscheduling strategies makes them a very promising solution to the scheduling needs of workstation clusters. In this paper we present a simulation study that compares, for a variety of workloads (that include both parallel and sequential applications) and operating system schedulers, 12 implicit coscheduling strategies in terms of the performance they are able to deliver to applications. By using a detailed simulator, we evaluate the performance of different coscheduling alternatives for a variety of simulation scenarios, and we identify the set of strategies that deliver the best performance to all the applications composing typical cluster workloads. Moreover, we show that for schedulers providing i>immediate preemption, the best strategies are also the simplest ones to implement.
UR - https://iris.uniupo.it/handle/11579/6757
U2 - 10.1023/A:1011416914729
DO - 10.1023/A:1011416914729
M3 - Article
SN - 1386-7857
VL - 4
SP - 121
EP - 131
JO - Cluster Computing
JF - Cluster Computing
IS - 2
ER -