Scalability analysis of embarassingly parallel applications on large clusters

Abstract

Abstract-This work presents a scalability analysis of embarrassingly parallel applications running on cluster and multi-cluster machines. Several applications can be included in this category. Examples are Bag-of-tasks (BoT) applications and some classes of online web services, such as index processing in online web search. The analysis presented here is divided in two parts: first, the impact of front end topology on scalability is assessed through a lower bound analysis. In a second step several task mapping strategies are compared from the scalability standpoint.

Large-scale branch contingency analysis through master/slave parallel computing

“…Reference [12] discussed static and dynamic parallel scheduling for execution times of various tasks underlying a heterogeneous platform. Recently, a ''superior/master/slave'' hierarchical topology for task assignment is presented in [13]. Reference [13] revisited the classic master/slave solution to schedule embarrassingly parallel tasks, which is the master for scheduling tasks heuristically.…”

“…Recently, a ''superior/master/slave'' hierarchical topology for task assignment is presented in [13]. Reference [13] revisited the classic master/slave solution to schedule embarrassingly parallel tasks, which is the master for scheduling tasks heuristically. Various task-process mapping algorithms are studied according to input data patterns.…”