Some Embedded systems involve multiple processors connected by a communication medium. These processors communicate with each other exchanging short messages using some message based communication protocols such as Controller Area Network (CAN), SAE J1939 etc. With the induction of new features inside such networks, the number of messages will increase. The increase of messages may affect the n...

This paper presents three examples of real-time image processing applications that were implemented on a data-ow architecture developed at the ETCA. Low-level image processing is performed on a regular three-dimensional network of 1024 custom data-ow processors. Image features previously extracted in the low-level step are handled by a two-dimensional network of 12 general purpose processors. F...

Functional or dataflow models of computationenable a program’s run-time system to determine which portions of it must be repeated when faults occur. Modifications were made to the run-time system of Cilk a threaded C to enable a Network of Workstations to tolerate fail-stop faults of individual processors or the network. The overheads needed to provide this fault tolerance are shown to be mainl...

ÐPresently, massively parallel processors (MPPs) are available only in a few commercial models. A sequence of three ASCI Teraflops MPPs has appeared before the new millenium. This paper evaluates six MPP systems through STAP benchmark experiments. The STAP is a radar signal processing benchmark which exploits regularly structured SPMD data parallelism. We reveal the resource scaling effects on ...

The paper investigates deterministic and randomized protocols for achieving broadcast (distributing a message from a source to all other nodes) in arbitrary multi-hop synchronous radio networks. The model consists of an arbitrary (undirected) network, with processors communicating in synchronous time-slots subject to the following rules. In each time-slot, each processor acts either as a transm...

Two algorithms for LU{decomposition on a transputer based reconngurable MIMD parallel computer with distributed memory have been analyzed in view of the interdependence of granularity and execution time. In order to investigate this experimentally, LU{decomposition algorithms have been implemented on a parallel computer, the Parsytec SuperCluster 128. The results of this investigation may be su...

The Multiswitch project at Purdue University is investigating broadcast techniques for use in point-to-point networks. This paper analyzes four techniques for delivery of a broadcast message to all processors in a network. The current Multi-switch prototype network uses one of these techniques, called Flooding with Duplicate Discard (FDD). FDD is attractive because it provides high reliability ...

Ÿ¡ S¢s£'¤ ¥S¦§£z ̈ XXX CAN WE USE A DIFFERENT TITLE THAT FORMATS BETTER? E.G. ADAPTIVE BSP ON A NETWORK OF WORKSTAIONS? XXX Several computing environments, including wide area networks and non-dedicated networks of workstations, are characterized by frequent unavailability of the participating machines. Parallel computations, with interdependencies among their component processes, cannot make pr...

ÐFully populated torus-connected networks, where every node has a processor attached, do not scale well since load on edges increases superlinearly with network size under heavy communication, resulting in a degradation in network throughput. In a partially populated network, processors occupy a subset of available nodes and a routing algorithm is specified among the processors placed. Analogou...