Scalable Visualization using Commodity Clusters: 
Challenges and Solutions

James T. Klosowski




In recent years, the performance of commodity computer components has improved dramatically.  Processors, graphics adapters, and network adapters, for example, have all exhibited significant improvements while maintaining a reasonable cost.  Due to the increase in the price/performance ratio of computers utilizing such components, clusters of commodity machines have become commonplace in today's computing world and are steadily displacing specialized, high-end, shared-memory machines for many graphics and visualization workloads.

Utilization and acceptance of commodity clusters has been a slow process at times due to the significant challenges introduced when switching from a share-memory architecture to a distributed memory one.  In this presentation, I will discuss these challenges and the many solutions that have been developed in recent years.  At the forefront will be the issue of scalability, whether it be scaling pixels, date, or overall performance.  As the nature of commodity hardware components suggests, the solutions are largely software-based, and include middleware layers for distributing the graphics workload across the cluster as well as for aggregating the final results to display for the user.