A New Compacting Non-Contiguous Processor Allocation Algorithm for 2D Mesh Multicomputers

Abstract

In non-contiguous allocation, a job request can be split into smaller parts that are allocated possibly non-adjacent free sub-meshes rather than always waiting until a single sub-mesh of the requested size and shape is available. Lifting the contiguity condition is expected to reduce processor fragmentation and increase system utilization. However, the distances traversed by messages can be long, and as a result the communication overhead, especially contention, is likely to increase. The extra communication overhead depends on how the allocation request is partitioned and assigned to free sub-meshes. In this paper, a new non-contiguous processor allocation strategy, referred to as Compacting Non-Contiguous Processor Allocation Strategy (CNCPA), is suggested for the 2D mesh multicomputers. In the proposed strategy, a job is compacted into free locations. The selection of the free locations has for goal leaving large free sub-meshes in the system. To evaluate the performance improvement achieved by the proposed strategy and compare it against well-known existing non-contiguous allocation strategies, the authors conducted extensive simulation experiments. The results show that the proposed strategy can improve performance in terms of job turnaround times and system utilization.

An Efficient All Shapes Busy List Processor Allocation Algorithm for 3D Mesh Multicomputers

“…Extensive simulation experiments have been conducted to study the effect of all request shapes approach on the performance of the BL allocation strategy. As in previous research (Yoo & Das, 2002;Chang & Mohapatra, 1998;Chiu & Chen, 1999;Ababneh, 2006;Ababneh, 2008;Ababneh, 2001;Ababneh, 2009;Ababneh et al, 2015;Ababneh et al, 2010;Bani-Mohammad & Ababneh, 2013;Bani-Mohammad et al, 2010;Bani-Mohammad et al, 2011;Bani-Mohammad et al, 2015;Bani-Mohammad et al, 2007a;Bani-Mohammad et al, 2006;Bani-Mohammad et al, 2007b;Bani-Mohammad et al, 2009;Lo et al, 1997;Zhu, 1992), the main system performance parameters used are the average turnaround time of jobs and mean system utilization, where the turnaround time of a job is the time that the job spends in the mesh system from arrival to departure, while the system utilization is the percentage of processors that are utilized over time. The important independent variable in the simulation is the system load.…”

“…It is defined as the inverse of the mean inter-arrival time of jobs. Its range of values, from low to heavy loads, has been determined through experimentation with the simulator, allowing each allocation strategy to reach its upper limits of utilization (Ababneh, 2006;Ababneh, 2008;Ababneh, 2001;Ababneh, 2009;Ababneh et al, 2015;Ababneh et al, 2010;Bani-Mohammad & Ababneh, 2013;Bani-Mohammad et al, 2010;Bani-Mohammad et al, 2011;Bani-Mohammad et al, 2015;Bani-Mohammad et al, 2007a;Bani-Mohammad et al, 2006;Bani-Mohammad et al, 2007b;Bani-Mohammad et al, 2009;Lo et al, 1997;Zhu, 1992).…”

“…In the simulation experiments, the synthetic workload models have been considered, where the synthetic workload models used in this research paper are based on the models used in previous research studies (Chang & Mohapatra, 1998;Chiu & Chen, 1999;Choo et al, 2000;Ababneh, 2006;Ababneh, 2008;Ababneh, 2001;Ababneh, 2009;Ababneh et al, 2015;Ababneh et al, 2010;Bani-Mohammad & Ababneh, 2013;Bani-Mohammad et al, 2010;Bani-Mohammad et al, 2011;Bani-Mohammad et al, 2015;Bani-Mohammad et al, 2007a;Bani-Mohammad et al, 2006;Bani-Mohammad et al, 2007b;Bani-Mohammad et al, 2009;Lo et al, 1997;Zhu, 1992).…”

“…Thus, the number of allocated and free sub-meshes is larger in this case, which means that the probability of successful allocation is increased. These distributions have often been used in the literature (Chiu & Chen, 1999;Ababneh, 2009;Ababneh et al, 2015;Ababneh et al, 2010;Bani-Mohammad & Ababneh, 2013;Bani-Mohammad et al, 2015).…”

“…The previous studies on allocation (Chiu & Chen, 1999;Choo et al, 2000;Ababneh, 2006;Lo et al, 1997) have shown that contiguous allocation is prone to low system utilization. Accordingly, non-contiguous allocation has been suggested with the goal of increasing system utilization by allowing dispersed free processors to be allocated to a parallel job (Chang & Mohapatra, 1998;Ababneh, 2008;Bani-Mohammad et al, 2015;Bani-Mohammad et al, 2007a;Lo et al, 1997) instead of always waiting until a single sub-mesh of the requested size and shape is available.…”