Sugath Warnakulasuriya

Learn More
Deadlock-free routing algorithms have been developed recently without fully understanding the frequency and characteristics of deadlocks. Using a simulator capable of true deadlock detection, we measure a network’s susceptibility to deadlock due to various design parameters. The effects of bidirectionality, routing adaptivity, virtual channels, buffer size(More)
Deadlock avoidance-based and deadlock recovery-based routing algorithms have been proposed in recent years without full understanding of the likelihood and characteristics of actual deadlocks in interconnection networks. This work models the interrelationships between routing freedom, message blocking, correlated resource dependencies and deadlock(More)
This work characterizes how various network parameters influence message blocking and deadlocks in irregular networks. Information on blocking behavior is provided that is useful in making design trade-offs between restricting routing freedom and allowing the possibility for deadlocks to form in irregular networks. This paper also identifies ways in which a(More)
We develop and present an approach to modeling resource allocations and dependencies within cut-through and wormhole interconnection networks. We also present a theoretical framework which allows the precise description of various types of message blocking behavior including deadlock. We show that a knot within a channel wait-for graph is a necessary and(More)
For uniprocessor systems, the object-oriented approach to programming has proven to be effective in obtaining highly maintainable and reusable software. Object-oriented programming languages, with its inherent conceptual elegance, has also allowed a high degree of programmability in the development of complex systems. As we move towards addressing issues(More)
DASH is a scalable shared memory multiprocessor architecture which employs directory based cache coherence. Due to the physical distribution of the memory in DASH, there is potential for long memory latency. Although a number of latency reducing and hiding techniques such caching, a weaker memory consistency model, and pre-fetching are employed by DASH, the(More)
  • 1