Title :
Evaluating the Performance of Parallel Computer System Using Recursive Models
Author :
Oguike, O.E. ; Agu, M.N. ; Echezona, S.C. ; Ebem, D.U.
Author_Institution :
Dept. of Comput. Sci., Univ. of Nigeria, Nsukka, Nigeria
Abstract :
A shared memory parallel computer system can be considered as a parallel computer system with a shared single ready queue. This can depict a single queue parallel server queuing model. Some models that are based on probability density function have been used to model the performance of the queuing system. This paper uses recursive to evaluate the performance of a single queue parallel server queuing model of compute intensive applications of a parallel computer system. The recursive models that this paper uses are efficient models because each recursive model makes one recursive call.
Keywords :
parallel processing; probability; queueing theory; recursive functions; shared memory systems; parallel computer system; parallel server; probability density function; queuing model; recursive model; shared memory system; compute intensive applicationr; parallel computer queuing systemt; performance metrics; recursive models;
Conference_Titel :
Computer Modeling and Simulation (EMS), 2010 Fourth UKSim European Symposium on
Conference_Location :
Pisa
Print_ISBN :
978-1-4244-9313-5
Electronic_ISBN :
978-0-7695-4308-6
DOI :
10.1109/EMS.2010.85