Graham P. Phillips
Computer Science Department
University of Southern California
Los Angeles, CA 90098-0781
graham@isi.edu
The decreasing cost of personal workstations over recent years has led to their proliferation in computing environments. Studies have shown that personal workstations are under-utilized for long periods of time during regular working hours. This paper identifies this under-utilization as the idle-workstation problem.
A taxonomy of computational models of systems that have attempted to solve the idle workstation problem is presented. The computational model includes issues such as, whether the user view of resources is location-transparent, whether heterogeneous processor instruction sets are supported, whether systems can interoperate and whether job migration is supported. The paper also presents those features that designers of future distributed systems should address to motivate users to pool their workstations.