Dynamic scheduling of process groups

In this paper we introduce the concept. of temporal locality of communication [or process groups and a hierarchical decision model [or dynamic scheduling of process groups. Empirical evidence suggests that, once a member of a process group starts to communicate with other processes in the group, it will continue to do so, while an independent process will maintain its state of isolation for some time. Other instances of inertial behavior of programs arc known. Temporal and spatial locality of reference are example of inertial behavior of programs, exploited by hierarchical storage systems; once a block of information (program or data) is brought into faster storage, it. is very likely that It will be referenced again within a short time frame. When process groups exhibit temporal locality of communication, this information can be used to hide the latency of paging and I/O operations, to perform dynamic scheduling to reduce processor fragmentation, and to identify optimal instances of time for check-pointing of process groups. In our scheduling model the supervisory process of a process group collects information about the dynamics of the group and shares it with local and global scheduling agents . • Work supported in part by NSF grants BlR-9301210 and MCR-9527131, by the Scalable 1/0 Initiative, by a grant from Intel Corporation and by CNPq Brazil