such as instruc-tions. Each stream is independent of all other streams, and each element of a stream can consist of one or more ob-jects or actions. We thus have four com-binations that describe most familiar parallel architectures: (1) SISD: single instruction, single data

We propose a hardware-assisted memory snapshot to improve software concurrency. It is built on top of the hardware resources for transactional memory and allows for easy development of system software modules such as concurrent garbage collector and dynamic profiler.

This paper proposes Update Plans as a specification formalism for abstract machines for parallel architectures. Update Plans are a formal specification language for abstract and concrete machines. First results in using Update Plans to specify parallel architectures are illustrated, and some

PROTEUS is a high-performance simulator for MIMD multiprocessors. It is fast, accurate, and flexible: it is one to two orders of magnitude faster than comparable simulators, it can reproduce results from real multiprocessors, and it is easily configured to simulate a wide range of architectures...

We survey general techniques and open problems in numerical linear algebra on parallel architectures. We first discuss basic principles of parallel processing, describing the costs of basic operations on parallel machines, including general principles for constructing efficient algorithms. We

of parallel architectures. It can be used as a tool for architectural level design space exploration of BSP computers both in research and teaching. The paper introduces BSP, BSPlib and BSPlab. Then it presents the architectural models and their parameters, and discusses how they can be used to make

In this report, we present different classification approaches for parallel architectures. Up to now, no general classification scheme is known. Our description of the various architectures does focus on the problem of trying to develop a rigorous classification system. There are many attempts

We theoretically and experimentally analyse the efficiency with which a wide range of important scientific computations can be performed on bulk synchronous parallel architectures.

. The scalable implementation of multigrid methods for machines with several thousands of processors is investigated. Parallel performance models are presented for two dierent structured-grid multigrid algorithms. These performance models are then used in the discussion of two implementation topics

This thesis describes techniques for the design of parallel programs that solvewell-structured problems with inherent symmetry. Part I demonstrates the reduction of such problems to generalized matrix multiplication by a group-equivariant matrix. Fast techniques for this multiplication