Dynamic load balancing through process and read-site placement in a distributed system

This paper explains various methods for increasing performance in a distributed system. The focus of this study is load balancing and its relation to optimal process and read-site placement. The system model is based on the LOCUS distributed file system, which allows replicated files. The simulation system model includes process migration, and the CSS (centralized synchronization site) program enforces a synchronization policy. All requests to open a file for access must be sent to the file's CSS, which checks for access conflicts. An algorithm is provided that increases system performance through load balancing, basing its decisions on data collected by the system. We analyze and discuss the algorithm's characteristics and effects on system performance. Results show the load-balancing algorithm, if properly tuned, can improve performance.