This report presents a model-driven, stress test methodology aimed at increasing chances of discovering faults related to network traffic in Distributed Real-Time Systems (DRTS). T...
Using a single traditional gang scheduling algorithm cannot provide the best performance for all workloads and parallel architectures. A solution for this problem is the use of...
Message Sequence Charts (MSCs) are widely used for describing interaction scenarios between the components of a distributed system. Consequently, worst-case response time estimati...
Markets of computing resources typically consist of a cluster (or a multi-cluster) and jobs that arrive over time and request computing resources in exchange for payment. In this p...
Sergei Shudler, Lior Amar, Amnon Barak, Ahuva Mu'a...
As we continue to evolve into large-scale parallel systems, many of them employing hundreds of computing engines to take on mission-critical roles, it is crucial to design those s...
Yanyong Zhang, Mark S. Squillante, Anand Sivasubra...