Improvements in Gang Scheduling for Parallel Supercomputers

Fabricio Alves Barbosa da Silva (1) , Miguel Campos , Isaac Scherson

1 ASIM - Architecture des Systèmes intégrés et Micro électronique

Gang scheduling has been widely used as a practical solution to the dynamic parallel job scheduling problem. Parallel threads of a single job are scheduled for simultaneous execution on a parallel computer even if the job does not fully utilize all available processors. Non allocated processors go idle for the duration of the time quantum assigned to the threads. In this paper we propose a class of scheduling policies, dubbed Concurrent Gang, that is a generalization of gangscheduling, and allows for the flexible simultaneous scheduling of multiple parallel jobs, thus improving the space sharing characteristics of gang scheduling. However, all the advantages of gang scheduling such as responsiveness, efficient sharing of resources, ease of programming, etc., are maintained. The resulting policy is simulated and compared with gang scheduling using a general purpose event driven simulator specially developed for this purpose.

Domaines: Informatique [cs]

Fabricio Alves Barbosa da Silva, Miguel Campos, Isaac Scherson. Improvements in Gang Scheduling for Parallel Supercomputers. 8th International Parallel Computing Workshop, Sep 1998, Singapore, Singapore. ⟨hal-01623367⟩