Parle '94 Parallel Architectures and Languages Europe: 6th International Parle Conference, Athens, Greece, July 4 - 8, 1994. Proceedings

ISBN-13: 9783540581840 / Angielski / Miękka / 1994 / 843 str.

Costas Halatsis; Dimitrios Maritsas; George Philokyprou

Opis

This volume features 60 papers and 24 poster presentations from a conference on parallel architectures, covering the entire field of parallel computing and providing a state-of-the-art report on research and advanced applications.

Spis treści

Improved probabilistic routing on generalized hypercubes.- Efficient data communication in Incomplete Hypercube.- Efficient communication in the folded Petersen interconnection networks.- Compiling rewriting onto SIMD and MIMD/SIMD machines.- A compilation technique for varying communication cost NUMA architectures.- A data partitioning algorithm for distributed memory compilation.- Towards a high precision massively parallel computer.- On the testability of purely recursive digital filters.- Voting on synchronous communications.- Protocol validation: A parallel technique to reduce the reachability tree.- Optimal parallel execution of complete binary trees and grids into most popular interconnection networks.- Time-optimal nearest-neighbor computations on enhanced meshes.- Simulation of PRAM models on meshes.- On the multisearching problem for hypercubes.- Code Liberation — A tool for refitting code to a parallel environment.- High-level visualisation of parallel logic program execution.- Next-generation parallel performance visualization: A prototyping environment for visualization development.- A formal approach to modeling expected behavior in parallel program visualizations.- An improvement in the size-depth tradeoff for strictly non-blocking generalized-concentration networks.- Bypass strategy in hypercube multiprocessors by adding bus connections.- Optimul: A hybrid multiprocessor for distributed and shared memory multiprocessing with WDM optical fiber interconnections.- ILIAS, a sequential language for parallel matrix computations.- Static communications in parallel scientific programs.- Load balancing for extrapolation methods on distributed memory multiprocessors.- On the definition of speedup.- Petri net performance models of parallel systems — Methodology and case study.- Run-time optimization of sparse matrix-vector multiplication on SIMD machines.- A new approach to array redistribution: Strip mining redistribution.- Analytic model of a Cache Only Memory Architecture.- Memory assignment for multiprocessor caches through grey coloring.- An adaptive update-based cache coherence protocol for reduction of miss rate and traffic.- Eliminating invalidation in coherent-cache parallel graph reduction.- Enhancing parallelism by removing cyclic data dependencies.- An interaction-based language and its typing system.- On the parallelism of data.- Predicative polymorphism in ?-calculus.- Improving the execution of the dependent and-parallel Prolog DDAS.- OR-parallel Prolog on distributed memory systems.- Efficient distributed memory implementation of a data parallel functional language.- Modelling photochemical pollution using parallel and distributed computing platforms.- Parallel algorithms and interval selection strategies for globally adaptive quadrature.- Systolic designs for evaluating linear combinations of Chebyshev polynomials.- The impact of the KSR1 allcache architecture on the behaviour of the DBS3 parallel DBMS.- Sparse Generalized HyperGrids for performance scalability.- A distributed garbage collector for active objects.- Bus based parallel computers: A viable way for massive parallelism.- Implementation of GTS.- A hierarchical activation management technique for fine-grain multithreaded execution.- Scheduling task graphs onto distributed memory multiprocessors under realistic constraints.- A dynamic algorithm for online scheduling of parallel processes.- CoReA: A synchronous calculus of parallel communicating reactive automata.- Deriving histories of nets with priority relation.- What do you get from a ?-calculus semantics?.- Rewriting: An effective model of concurrency.- Mercury: Object-affinity scheduling and continuation passing on multiprocessors.- Dynamic allocation of communication bandwidth in multicomputers.- Load balanced priority queues on distributed memory machines.- Parallel computations with algebraic numbers a case study: Jordan normal form of matrices.- Batch dynamic algorithms for two graph problems.- A general method for maximizing the error-detecting ability of distributed algorithms.- MAGE: A Massively Agent Execution System for logic programming.- Checking the serial correctness of control-parallel programs.- NXLib — A parallel programming environment for workstation clusters.- Applying human factors techniques to the design of a tool for parallel program performance analysis and tuning.- An Actor based programming system for heterogeneous processing.- Introducing fairness in CSP systems specified with state variables.- A dataflow-based massively parallel programming language and its implementation.- Using subpages for coherency control in parallel database systems.- A new multi-dimensional interconnection network for SIMD architectures.- A scalable bit-sequential SIMD architecture for pattern recognition.- An iterative array processor architecture for matrix computation.- Datarol-II: A fine-grain massively parallel architecture.- From BSP to a virtual von Neumann machine.- Data parallel relational database processing on the connection machine CM-2 with DataVault.- Staggered scheme: A loop allocation policy.- Using parallel simulated annealing in the mapping problem.- A new parallel MIMD Connected Component Labeling algorithm.- Using reference counters in update based coherent memory.- Compositional reasoning about real-time distributed systems with limited resources.- Dynamic load balancing on a virtually-shared memory parallel computer system.- Analysing the load balancing scheme of a parallel system on multiprocessors — A modeling approach —.- Performance of interconnection network in multithreaded architectures.- Scheduling algorithms performance with the pSystem parallel programming environment.- A new routing scheme for concatenating two omega networks.