Gregory D Peterson and Roger D Chamberlain 1996 Distrib. Syst. Engng. 3 9 doi:10.1088/0967-1846/3/1/003
Gregory D Peterson and Roger D Chamberlain
Show affiliationsThe utilization of networked, shared, heterogeneous workstations as an inexpensive parallel computational platform is an appealing idea. However, most performance models for parallel computation are oriented towards the use of tightly-coupled, dedicated, homogeneous processors. We develop and validate an analytic performance modelling methodology for synchronous iterative algorithms executing on networked workstations. The model includes the effects of application load, background load, and processor heterogeneity. We use two applications, nonlinear optimization and discrete-event simulation, to validate the model. Various policies for the use of the workstations are considered and the optimal (or near-optimal) scheduling found. The performance modelling methodology provides significant help in addressing scheduling and similar issues in a shared resource environment.
07.05.Bx Computer systems: hardware, operating systems, computer languages, and utilities
02.60.Pn Numerical optimization
84.40.Ua Telecommunications: signal transmission and processing; communication satellites
Issue 1 (March 1996)
Gregory D Peterson and Roger D Chamberlain 1996 Distrib. Syst. Engng. 3 9
Jeffrey S. Bary et al 2009 ApJ 706 L168
Martin Cohen et al. 1999 The Astronomical Journal 117 1864
J Hough 2007 J. Phys.: Conf. Ser. 66 012002
C N Pope and L J Romans 1990 Class. Quantum Grav. 7 97
Amokrane Berdja et al 2006 J. Opt. A: Pure Appl. Opt. 8 244
M de Roo et al 1990 Class. Quantum Grav. 7 1181
Adam Muzzin et al 2009 ApJ 706 L188
B N Taylor and P J Mohr 1999 Metrologia 36 63
Brian K Canfield et al 2006 J. Opt. A: Pure Appl. Opt. 8 S278