Load Balancing White Papers

Efficient Load Balancing for Wide-Area Divide-and-Conquer Applications

Overview Divide-and-conquer programs are easily parallelized by letting the programmer annotate potential parallelism in the form of spawn and sync constructs. To achieve efficient program execution, the generated work load has to be balanced evenly among the available CPUs. For single cluster systems, Random Stealing (RS) is known to achieve optimal load balancing. However, RS is inefficient when applied to hierarchical wide-area systems where multiple clusters are connected via Wide-Area Networks (WANs) with high latency and low bandwidth. This paper experimentally compares RS with existing load-balancing strategies that are believed to be efficient for multi-cluster systems, Random Pushing and two variants of Hierarchical Stealing. It demonstrates that, in practice, they gain less than optimal results.

Further White Paper Details
PublisherAssociation for Computing Machinery File FormatPDF
Date PublishedJanuary 2008
FormatWhite Papers   
Topics

Quick Sitemap Links: