The main objective of this paper is to put forward a software process model for high-performance systems (HPS), and to present a formal framework to describe software design methodologies (SDMs) for those systems. The framework consists of two main parts: the software process activities which characterise the development of HPS, and the components of the SDM (concepts, artifacts, representation and actions) which are essential for any methodology. The framework relates these two parts by identifying generic components of each activity in the software process that can be used to classify and evaluate SDMs for HPS. The framework has been formally specified using the language Z and used to derive formal specifications of SDMs. This is illustrated in the paper by presenting part of the specification of oDM (an occam design method). 1 Introduction The number of software design methodologies (SDMs) currently available is enormous. This means that designers and managers are faced with the d...

Abstract not found

The performance of logical process based parallel simulation (PS) protocols like Time Warp and Chandy/Misra/Bryant is influenced by a variety of factors such as the event structure underlying in the simulation model, the partitioning into submodels, the performance characteristics of the execution platform, the implementation of the simulation engine and optimizations related to the protocols. The mutual performance effects of parameters exhibit a prohibitively complex degree of interweaving, giving analytical performance investigations only relative importance. Nevertheless, performance analysis is of utmost practical interest for the simulationist who wants to decide on the suitability of a certain PS protocol for a specific simulation model before substantial efforts are invested in developing sophisticated PS codes. Since PS performance prediction based on analytical models appears doubtful with respect to adequacy and accuracy, this work presents a prediction methodology based on ...

Most of the methods proposed for the development of high-performance systems (HPS) do not balance the software and performance engineering activities. This paper presents a method for the development of HPS which promotes the production of well-engineered, highly parallel programs the design decisions of which are also guided by their impact on performance. The method follows strong software engineering principles such as modularity, and the use of formal methods to support verification and transformation. From the point of view of performance engineering the method supports the use of prototypes for performance prediction at early stages of the development. Much of the method is language and hardware independent, and the paper illustrates the application of the method to the development of occam programs running on transputers. 1

ion Graph (SAG) describes a machine as a rooted tree composed of System Abstraction Unit (SAU). A SAU is a tuple with four components (P=Processing, M=Memory, C/S=Communication/Synchronisation and I/O=Input/Output). Each level of the SAG characterises an aspect of the machine, using one or several components of the SAU. Advantages and Drawbacks On the one hand, this linear model is simple to implement. Besides, once a machine has been characterised, its model can be used by any program. It means that if p programs are combined with m machines there will only be p+m characterisations for all the combination. As a matter of fact, each machine has to be characterised once and also each program has to be analysed once. In addition, it is accurate enough to get an insight into the performance of a specific program on a specific machine. On the other hand, the assumption of the linearity of execution is false. Indeed, even in a common desktop there is parallelism and also pipelining to exe...

The performance of logical process based distributed simulation (DS) protocols like Time Warp and Chandy/Misra/Bryant is influenced by a variety of factors such as the event structure underlying in the simulation model, the partitioning into submodels, the performance characteristics of the execution platform, the implementation of the simulation engine and optimizations related to the protocols. The mutual performance effects of parameters exhibit a prohibitively complex degree of interweaving, giving analytical performance investigations only relative importance. Nevertheless, performance analysis is of utmost practical interest for the simulationist who wants to decide on the suitability of a certain DS protocol for a specific simulation model before substantial efforts are invested in developing sophisticated DS codes. Since DS performance...

In this paper, we discuss the problem of developing performance-oriented software and the need for methodologies. We then present the EDPEPPS (Environment for Design and Performance Evaluation of Portable Parallel Software) approach to the problem of designing and evaluating high-performance (parallel) applications. The EDPEPPS toolset is based on a rapid prototyping philosophy, where the designer synthesises a model of the intended software which may be simulated, and the performance is subsequently analysed using visualisation. The toolset combines a graphical design tool, a simulation facility, and a visualisation tool. The same design is used to produce a code suitable for simulation and real execution. 1 Introduction Many existing software engineering techniques which assist in the management of the complexity of serial programs can be directly applied to the development of parallel software. Many of the assumptions about serial programs, however, do not hold for parallel programs...

The results presented here are based on many years of experience of development and application of DYANA – an environment for analysis of multiprocessor computer systems operation. The architecture and basic features of such an environments are discussed. Main problems arizing during such an environments design are highlighted and possible solutions are shown. The key features of the DYANA environment are: the possibility of both quantitative and algorithmic analysis of system to be modeled; the time complexity estimation subsystem which helps to avoid the instruction-level simulation of target computer system; support of program development through simulation. 1.

This paper describes the architecture of a development environment for computer-aided parallel software engineering. The environment comprises tools for program design, simulation, and visualisation. The toolset supports an interactive, performance-oriented software development life-cycle, and the user interface makes extensive use of graphics and animation. 1 Introduction The recent rapid growth of parallel cluster computing has signalled the urgent need for tools to support a parallel program design life-cycle. Concurrent programs are certainly no less complex than serial ones, but whereas serial programmers can rely on a simple and stable programming model (the von Neumann machine), parallel programmers must cope with two additional issues: platform configuration; and mapping (of the concurrent program onto the parallel platform). In high-performance computing, many assumptions made by serial software engineers no longer hold. For example, in serial life-cycle models, performance e...

this report other projects with similar goals to EDPEPPS are reviewed and their strengths and weaknesses are highlighted. 2 The HAMLET Project - ESPRIT 6290