International Journal of Cloud Applications and Computing, 5(1), 1-13, January-March 2015 1
Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
ABSTRACT
This research paper presents a new contiguous allocation strategy for 3D mesh-connected multicomputers.
The proposed strategy maintains a list of maximal free sub-meshes and gives priority to allocating corner and
boundary free sub-meshes. The goal of corner and boundary allocation is to decrease the number of leftover
free sub-meshes and increase their sizes, which is expected to reduce processor fragmentation and improve
overall system performance. The proposed strategy, which is referred to as Turning Corner-Boundary Free
List (TCBFL) strategy, is compared, using extensive simulation experiments, to several existing allocation
strategies for 3D meshes. These are the First-Fit (FF), Turning First-Fit Free List (TFFFL), and Turning
Busy List (TBL) allocation strategies. The simulation results show that TCBFL produces average turnaround
times and mean system utilization values that are superior to those of previous strategies.
Corner-Boundary Processor
Allocation for 3D Mesh-
Connected Multicomputers
Ismail Ababneh, Department of Computer Science, Al al-Bayt University, Mafraq, Jordan
Saad Bani-Mohammad, Department of Computer Science, Al al-Bayt University, Mafraq,
Jordan
Motasem Al Smadi, Department of Computer Science, Al al-Bayt University, Mafraq, Jordan
Keywords: Contiguous Allocation, Processor Fragmentation, Simulation, System Utilization, Turnaround
Time
1. INTRODUCTION
Mesh interconnection networks have been
extensively employed in large-scale multi-
computers due to their structural regularity,
simplicity, ease of implementation and scal-
ability (Ababneh, 1995; Bani-Mohammad et
al., 2006; Bani-Mohammad et al., 2009; Bani-
Mohammad & Ababneh, 2013; Foster, 1995;
Kumar et al., 2003; Lo et al., 1997; Seo & Kim,
2003; Zhu, 1992).
Effective processor allocation and job
scheduling are critical if the full computational
power of multicomputers is to be exploited
properly (Choo et al., 2000; Yoo & Das, 2002).
Processor allocation is responsible for determin-
ing the set of processors on which a parallel
job is executed, whereas job scheduling is re-
sponsible for selecting the order in which jobs
are selected for execution (Choo et al., 2000;
De Rose et al., 2007; Yoo & Das, 2002). The
job scheduler selects the next job to execute,
DOI: 10.4018/ijcac.2015010101