HELIOS: Hierarchical performances evaluation modelling of large information open systems

• Unité de recherche: PCRD EU
• Numéro de projet: 96.0318
• Titre du projet anglais: HELIOS: Hierarchical performances evaluation modelling of large information open systems

Mots-clé (Anglais)

Performance prediction; modeling; simulation; open systems; client/server; object oriented technologies; workbench

Autre Numéro de projet (Anglais)

EU project number: EP 22.354

Programme de recherche (Anglais)

EU-programme: 4. Frame Research Programme - 1.3 Telematic systems

Description succincte (Anglais)

See abstract

Partenaires et organisations internationales (Anglais)

IFATEC St quentin en Yvelines (F), Ecole Polytechnique Fédérale de Lausanne (CH),
EURIWARE St quentin en Yvelines (F), IONA Dublin (IRL), TRANSTOOLS'Madrid (E), Universitat de Les Illes Balears, Palma de Majorque (E), University of Southampton (GB)

Résumé des résultats (Abstract) (Anglais)

The objective of the project was to allow enterprises to cope with the wave of second generation client/server by helping the user to visualize and understand the way his/her workload would be handled by new middleware technologies. The next generation of client/server systems will inevitably be built using three-tiered architecture, where object oriented technology will be the ultimate form of self managing distribution.
The project has developed a workbench to predict and analyze performance and thereby aid the sizing and the configuring of commercial systems by service suppliers, service integrators, client/server editors and customers as well. It includes :
· Performance prediction of a particular configuration and workload,
· The ability to analyze and display information in a way easy to understand and which highlights bottlenecks
· Tuning assistance for an existing configuration for a particular workload.
As companies deploy mission-critical applications across distributed client/server environments, performance complexity becomes a major obstacle. Today's distributed environments are not only more technologically diverse and geographically dispersed than in the past, they are also more dynamic. Managing performance across many application, system and network components in these new environments is absolutely vital to minimize the impact to the business.
Modeling an Information System (IS) is a complex and cost consuming task. HELIOS approach consists in offering a wide library of predefined models for main computing products (hardware, software) in a workbench. The user-friendly interface of this tool and the re-usability of component models allows IS designers to get a fast modeling and a cheap performance prediction.
HELIOS work has been built on the background of SWAP/IFATEC (a simulation predictor), MultiBase/TransTOOLS' (a 4GL) and ORBIX/IONA ( an ORB). SWAP has proved the feasibility of the approach in working out a workbench (SMART) for a first generation client/server (ORACLE7).
The tangible results of the project are:
· A collection of performance models of Distributed Information System (DIS) components, CORBA/ORBIX, COM/OLE and Stored Procedures/PL-SQL.
· A performance modeling workbench for DIS designers to evaluate performance of alternative configurations.
· Tools for benchmarking, validating and adapting performance models.
· Demonstration of the validity of the workbench using real applications.

Références bases de données (Anglais)

Swiss Database: Euro-DB of the
State Secretariat for Education and Research
Hallwylstrasse 4
CH-3003 Berne, Switzerland
Tel. [phone]
Swiss Project-Number: 96.0318