In this paper we present the first system that implements OpenMP on a network of shared-memory multiprocessors. this system enables the programmer to rely on a single, standard, shared-memory API for parallelization w...
详细信息
In this paper we present the first system that implements OpenMP on a network of shared-memory multiprocessors. this system enables the programmer to rely on a single, standard, shared-memory API for parallelization within a multiprocessor and between multiprocessors. It is implemented via a translator that converts OpenMP directives to appropriate calls to a modified version of the TreadMarks software distributed memory system (SDSM). In contrast to previous SDSM systems for SMPs, the modified TreadMarks uses POSIX threads for parallelism within an SMP node. this approach greatly simplifies the changes required to the SDSM in order to exploit the intra-node hardware shared memory. We present performance results for six applications (SPLASH-2 Barnes-Hut and Water; NAS 3D-FFT, SOR, TSP and MGS) running on an SP2 with four four-processor SMP nodes. A comparison between the threaded implementation and the original implementation of TreadMarks shows that using the hardware shared memory within an SMP node significantly reduces the amount of data and the number of messages transmitted between nodes, and consequently achieves speedups up to 30% better than the original versions. We also compare SDSM against message passing. Overall, the speedups of multithreaded TreadMarks programs are within 7-30% of the MPI versions.
the establishment of a low-cost, effective testing and validation strategy has been pursued by many researchers at the program level as well as at the specification level. the application of mutation testing for valid...
详细信息
the establishment of a low-cost, effective testing and validation strategy has been pursued by many researchers at the program level as well as at the specification level. the application of mutation testing for validating specifications based on statecharts is proposed. A mutation operator set for statecharts, one of the crucial points for effectively applying mutation testing is defined; in this scope these operators can be taken as a fault model. We also provide strategies to abstract the statechart components according to different statechart features that may comprise the testing and validation activity aims, providing in this way mechanisms for the establishment of an incremental, hierarchical, mutation-based testing strategy. Implementation and functional aspects of PROTEUM/ST, a tool under development are also presented.
the proceedings contain 31 papers. the special focus in this conference is on Static Analysis and Software Methodologies. the topics include: logicprogramming and model checking;the copying approach to tabling;a mult...
ISBN:
(纸本)3540650121
the proceedings contain 31 papers. the special focus in this conference is on Static Analysis and Software Methodologies. the topics include: logicprogramming and model checking;the copying approach to tabling;a multithreaded execution environment for sicstus prolog;a framework for bottom up specialisation of logic programs;termination of logic programs with block declarations running in several modes;the boolean logic of set sharing analysis;derivation of proof methods by abstract interpretation;detecting unsolvable queries for definite logic programs;staging static analyses using abstraction-based program specialization;type domains and type representations for logic programs;a methodology for language development;explicit substitutions for objects and functions;the complexity of late-binding in dynamic object-oriented languages;a compiler for rewrite programs in associative-commutative theories;solution to the problem of zantema on a persistent property of term rewriting systems;a general framework for r-unification problems;operational versus denotational methods in the semantics of higher order languages;functional implementations of continuous modeled animation;compiling erlang to scheme;compiling lazy functional programs for the java virtual machine;a game semantics foundation for logicprogramming;controlling search in declarative programs;encapsulating data in logicprogramming via categorical constraints;constructive negation using typed existence properties;optimal reduction of functional expressions;embedding multiset constraints into a lazy functionallogic language;a hidden herbrand theorem and integrating constraint propagation in complete solving of linear diophantine systems.
In recent works, we have proposed a general framework for lazy functionallogicprogramming with algebraic polymorphic types, i.e., parametric datatypes whose data constructors fulfill a given set of equational axioms...
详细信息
E-unification (i.e. solving equations modulo an equational theory E) is an essential technique in automated reasoning, functionallogicprogramming and symbolic constraint solving but, in general E-unification is unde...
详细信息
We introduce a semantics of logicprogramming based on an classical Game theory, which is proven to be sound and complete w.r.t. the traditional operational semantics and Negation as Failure. this game semantics is ba...
详细信息
In logicprogramming, almost no work has been done so far on proving that certain queries cannot succeed. Work in this direction could be useful for queries which seem to be non-terminating. Such queries are not excep...
详细信息
this paper deals withthe incremental detection of implicit equalities using the revised simplex method. this algorithm is more efficient and more suitable to practical problems than the tableau method usually applied...
详细信息
We define a framework for writing executable declarative specifications which incorporate categorical constraints on data, Horn Clauses and datatype specification over finite-product categories. We construct a generic...
详细信息
暂无评论