In this article we propose a “standard” performance tool interface for OpenMP, similar in spirit to the MPI profiling interface in its intent to define a clear and portable API that makes OpenMP execution events visible to performance libraries. When used together with the MPI profiling interface, it also allows tools to be built for hybrid applications that mix shared and distributed memory pr...

The current OpenMP 2.5 specification does not include a binding for the Java programming language. However, Java is a wide-spread programming language that is even used for HPC programming. We propose an adaptation of OpenMP to Java by retrofitting the basic OpenMP directives to Java and further propose some new concepts to make OpenMP fit into Java’s language philosophy. We discuss how Java’s ...

This paper describes compiler techniques that can translate standard OpenMP applications into code for distributed computer systems and optimization techniques to enhance the performance of realistic OpenMP application. OpenMP has emerged as an important model and language extension for shared-memory parallel programming. However, despite OpenMP's success on these platforms, it is not currently...

This paper presents a novel proposal to define task parallelism in OpenMP. Task parallelism has been lacking in the OpenMP language for a number of years already. As we show, this makes certain kinds of applications difficult to parallelize, inefficient or both. A subcommittee of the OpenMP language committee, with representatives from a number of organizations, prepared this proposal to give O...

A shortcoming of OpenMP standards to date is that they lack an application programming interface (API) to support construction of portable, efficient, and vendor-neutral performance tools. To address this issue, the tools working group of the OpenMP Language Committee has designed OMPT—a performance tools API for OpenMP. OMPT enables performance tools to gather useful performance information fr...

Mission critical applications, like those used in flight traffic control systems or in the reactive systems, need to be reliably executed to preserve integrity of their own operating environment and to ensure correct course of operations. The design of these systems must take into account aspects of reliability but, at the same time, cannot neglect the performance aspects. Techniques for adding...

Irregular algorithms are difficult to parallelize using existing OpenMP constructs. This paper concentrates on algorithms that deploy task pools, i.e., data structures for dynamic load balancing. We present several task pool variants that we have implemented in OpenMP, and compare their performance. Due to the lack of a mechanism in OpenMP to put a thread to sleep, we had to use busy waiting in...

CMP (Chip Multiprocessor) is a promising processor architecture, not only for high performance but also for reducing power and energy consumption in embedded applications. We have implemented an OpenMP compiler for an embedded Renesas M32R chip multiprocessor as a parallel programming environment. In this paper, we report the preliminary performance of OpenMP benchmarks, including scientific an...

We present a collection of C/C++ and Fortran programs with OpenMP directives that were designed to validate the correctness of an OpenMP implementation. The validation methodology and implemented tests are presented. We also discuss the differences between the Fortran and C validation suite and extensions made possible by the clarifications introduced by OpenMP 2.5.