A Model-Driven Performance Analysis Framework for Distributed, Performance-Sensitive Software Systems

  • Authors:

  • Swapna S. Gokhale;Aniruddha Gokhale;Jeffrey Gray

  • Affiliations:

  • University of Connecticut, Storrs, CT;Vanderbilt University, Nashville, TN;Univ. of Alabama at Birmingham

  • Venue:
  • IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 10 - Volume 11
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Large-scale, distributed, performance-sensitive software (DPSS) systems from the basis of mission- and often safetycritical applications. DPSS systems comprise of many independent artifacts, such as network/bus interconnects, many coordinated local and remote endsystems, and multiple layers of software. DPSS systems demand multiple, simultaneous predictable performance requirements such as end-to-end latencies, throughput, reliability and security, while also requiring the ability to control and adapt operating characteristics for applications with respect to such features as time, quantity of information, and quality of service (QoS) properties including predictability, controllability, and adaptability of operating characteristics for applications with respect to such features as time, quantity of information, accuracy, confidence and synchronization. All these issues become highly volatile in large-scale DPSS systems due to the dynamic interplay of the many interconnected parts that are often constructed from smaller parts.