provide tools or abstractions that allow developers to program in parallel. But what hardware do we need to support shared memory threads? The hardware should provide a well-defin...
We propose a new parallel topology discovery algorithm for irregular, mesh-connected networks with unidirectional links and wormhole routing. An algorithm of this type was develop...
Abstract. We analyze the convergence rate of an asynchronous space decomposition method for constrained convex minimization in a reflexive Banach space. This method includes as spe...
Compared to Krylov space methods based on orthogonal or oblique projection, the Chebyshev iteration does not require inner products and is therefore particularly suited for massiv...
Implicitcoscheduling is a distributed algorithm fortime-sharing communicating processes in a cluster of workstations. By observing and reacting to implicit information, local sche...
Andrea C. Arpaci-Dusseau, David E. Culler, Alan M....