There is growing recognition of the importance of the role of architecture in determining the quality of a software system. While a good architecture cannot guarantee attainment of quality goals, a poor architecture can prevent their achievement. It is particularly important to evaluate the performance of a distributed system architecture. Errors made early can cause excessive overhead for communication and coordination and they are far more difficult – if not impossible – to correct with tuning. This paper discusses assessment of the performance characteristics of distributed software architectures in early life cycle stages. The techniques are described and illustrated with a simple example. 1.

Abstract. Historically, high performance systems use schedulers and intelligent resource managers in order to optimize system usage and application performance. Most of the times, applications just issue requests of resources to the central system. This centralized approach is an unnecessary constraint for a class of potentially flexible applications, whose resource usage may be modulated as a function of the system status. In this paper we propose a tool which, in a way essentially transparent to final users, lets the application to self-tune in function of the status of the target execution environment. The approach hinges on the use of the MetaPL/HeSSE methodology, i.e., on the use of simulation to predict execution times and skeletal descriptions of the application to describe run-time resource usage. 1

This OMG document replaces the submission document (ad/01-06-14) and the draft adopted specification (ptc/02-01-20). It is an OMG Final Adopted Specification, which has been approved by the OMG board and technical plenaries, and is currently in the finalization phase. Comments on the content of this document are welcomed, and should be directed to issues@omg.org by May 2, 2002. You may view the pending issues for this specification from the OMG revision issues web page

This paper deals with the performance optimization of multiple-task applications in GRID environments. Typically such applications are launched by a Resource Manager, which only takes into account the application’s resource requirements and current availability on the GRID. Here a novel approach is presented, that performs resource management in user space, making it possible to exploit application modularity and flexibility and to take into account expected performance figures produced by GRID simulation. The objective is to make optimized choices that can lead to reduced application response times. After an introduction to the GRID simulation environment used, the structure of an application launcher able to optimize a number of application tasks and their mapping on the GRID is sketched, presenting the encouraging performance results obtained. 1

This document numbered ptc/2003-03-02 is the Draft Available Specification. It will form the basis for the formal publication of this standard upon completion of the adoption process. ptc/2003-03-02