Summary: Effective Distributed Scheduling of Parallel Workloads
Andrea C. Dusseau, Remzi H. Arpaci, and David E. Culler
Computer Science Division
University of California, Berkeley
fdusseau, remzi, cullerg@CS.Berkeley.EDU
Abstract
We present a distributed algorithm for time­sharing parallel
workloads that is competitive with coscheduling. Implicit
scheduling allows each local scheduler in the system to
make independent decisions that dynamically coordinate the
scheduling of cooperating processes across processors. Of
particular importance is the blocking algorithm which de­
cides the action of a process waiting for a communication or
synchronization event to complete. Through simulation of
bulk­synchronous parallel applications, we find that a sim­
ple two­phase fixed­spin blocking algorithm performs well;
a two­phase adaptive algorithm that gathers run­time data
on barrier wait­times performs slightly better. Our results
hold for a range of machine parameters and parallel program
characteristics. These findings are in direct contrast to the

Source: Arpaci-Dusseau, Remzi - Department of Computer Sciences, Department of Computer Sciences, University of Wisconsin at Madison

Collections: Computer Technologies and Information Sciences