The Cost of A BSP Algorithm
The cost of a superstep is determined as the sum of three terms; the cost of the longest running local computation, the cost of global communication between the processors, and the cost of the barrier synchronisation at the end of the superstep. The cost of one superstep for processors:
where is the cost for the local computation in process, and is the number of messages sent or received by process . Note that homogeneous processors are assumed here. It is more common for the expression to be written as where and are maxima. The cost of the algorithm then, is the sum of the costs of each superstep.
where is the number of supersteps.
, and are usually modelled as functions, that vary with problem size. These three characteristics of a BSP algorithm are usually described in terms of asymptotic notation, e.g. .
Read more about this topic: Bulk Synchronous Parallel
Famous quotes containing the word cost:
“If music in general is an imitation of history, opera in particular is an imitation of human willfulness; it is rooted in the fact that we not only have feelings but insist upon having them at whatever cost to ourselves.... The quality common to all the great operatic roles, e.g., Don Giovanni, Norma, Lucia, Tristan, Isolde, Brünnhilde, is that each of them is a passionate and willful state of being. In real life they would all be bores, even Don Giovanni.”
—W.H. (Wystan Hugh)