An Overview of the SimWorld Agent‐Based Grid Experimentation System
Matthias J. Scheutz, Jack J. Harris · 2011
This chapter provides an overview of the system architecture of SimWorld Agent-Based Grid Experimentation System (SWAGES). The new implementation of SWAGES in the distributed Agent Development Environment (ADE) is discussed, and some of the advantages of distributing its architecture are highlighted. The chapter describes an application of SWAGES in the context of a biological agent-based model to highlight how SWAGES addresses the challenges of exploring large model parameter spaces in high-performance computing (HPC) environments. Finally, it compares SWAGES to related large-scale simulation frameworks and summarizes its new features. In addition to parallelization, SWAGES also provides mechanisms to support asynchronous updating and scheduling within the simulation environment. SWAGES comprises several levels of error detection and recovery. At the infrastructure level, ADE provides component supervision, component monitoring, and component restarting in case components fail or crash. Controlled Vocabulary Terms error correction; smart grids; software agents