Distributed systems have become dramatically more significant with the advent of improved wired and wireless network technologies. There has been sustained academic interest in high-level techniques for constructing distributed systems, aiming to reduce development time and increase reliability. Collaborating with Motorola, we aim to evaluate the high-level techniques in a realistic distributed telecommunications context. The language platforms for our investigation are two distributed functional languages: Erlang is a simple language with proven telecommunications track record, GdH is a sophisticated research language that can be adapted by the research team.
The evaluation covers specific high-level techniques: high-level distributed coordination, sophisticated type systems, correctness, and non-strictness. Because these techniques abrogate control of low-level distribution issues we also investigate whether they can still deliver the required functionality, including such issues as soft real-time capabilities, fault-tolerance, and implementation scalability. Finally the project initiates the transfer of distributed functional language technology into the UK telecommunications sector.
Objectives
To assess the costs and benefits of high-level distributed programming
constructs for telecommunications software, and compare them with
conventional technology. Driving factors are time to market and
software quality, and the techniques we investigate are high level
distributed coordination, ease of validation/verification, impact of
sophisticated type systems and non-strictness.
To investigate whether high level programming paradigms can deliver telecommunications application functionality. Specific functionalities are foreign language interfaces, soft real-time capabilities, fault tolerance, resource reclamation, wireless LAN interfaces, implementation scalability and the scalability of the software development model.
To develop three medium-scale advanced telecommunications applications supplied by Motorola in two high-level functional languages: Erlang and GdH. The applications represent a range of telecommunications functionality and scale
- High-Level Technique for Distributed Telecommunications Software
- The project aims to evaluate distributed functional programming techniques for realistic telecommunications applications.
- ESPRC GR/R 88137
- Functional Distributed Interactive Systems
- The primary aim of the project is to study the suitability of distributed lazy functional langauges for constructing real applications with small-scale distribution.
- ESPRC GR/M 55633
- Evaluating Functional Distribution
- Evaluating two non-strict distributed functional languages, namely Eden, and GdH. People: Rita Loogen, Kevin Hammond.
- 4-year British Council/DAAD funded travel grant (Project No. 1097), with partners at Phillips-Universitaet Marburg, and St Andrews University.
- Performance Evaluation of Parallel Functional Programs for Multiprocessor Computer Systems
- The project aims to evaluate the performance of the Eden and GpH parallel functional programming languages, and to construct frameworks for reasoning about them. People: Ricardo Pena, Yolanda Ortega-Mallen, Kevin Hammond.
- 4-year British Council/Acciones Integradas travel grant (Project No. MDR 8472).
- Phil Trinder, Email:trinder@cee.hw.ac.uk, Lecturer.
- David King, Senior Software Engineer.
- Robert Pointon, Email:rpointon@cee.hw.ac.uk, Research Associate.
- Hans Wolfgang Loidl, Email:hwloidl@cee.hw.ac.uk, Research Fellow.