Probabilistically Analysable Real-Time Systems

Project Overview

Why is PROARTIS important?

Reliable computer systems such as aircraft flight control systems and engine control systems are of ever growing importance in the operation of aircraft and cars. These systems need to be developed to an exceptional level of reliability, yet their increasing complexity leads to enormous development cost, including the cost and time to test and analyze the system performance.

Timing techniques which predict with cycle accuracy the state of hardware and software through analysis cannot keep pace with these technological developments. Failure to develop new timing analysis techniques will hinder the development of future industrial applications and advanced safety systems in planes and cars.

Using probabilistic analysis is expected to promote dramatic performance improvements in, for example, the avionics industry. The modern aircraft requires millions of lines of code for on-board control functions such as guidance, navigation and control algorithms. Even in the most modest projections, a fourfold increase in performance will be required for on-board processors in next-generation aircraft. Existing hardware platforms may struggle to cope with these new demands.

This three-year project aspires to obtain research results that enable the use of complex processors in these types of systems, providing high performance in airplanes, cars and satellites and resulting in systems with more advanced safety and energy efficient features.

 

What will PROARTIS achieve?

PROARTIS will develop new tools, hardware and software architectures that will allow faster computer hardware features to be used and  analyzed more easily in reliable systems. The project will reduce the cost of performance analysis for new systems by 40% and help support the growth in complexity.

Demonstrating why pathological timing cases in complex software used in avionics and elsewhere only arise with negligible probability will ultimately permit more widespread and effective use of probabilistic analysis techniques in system verification and certification. 

PROARTIS will show that new advanced hardware/software features enabling truly randomized timing behaviour can be used in critical real-time embedded systems.

The project marks a radical break away from timing techniques that may lead to significantly increased performance and reliability in critical real-time embedded systems while reducing costs at the same time.

 

Who are the PROARTIS partners?

All partners have a recognised track record in the areas of expertise required to carry out the project.

Coordinated by the Barcelona Supercomputing Center, the PROARTIS Project brings together five top research centres such as the University of Padua (Italy), Institut National de Recherche en Informatique et Automatique (France) as well as industrial partners such as Rapita Systems (UK) and Airbus France (France). The project also collaborates with the University of Massachusetts Amherst (USA) and the University of York (UK).

The project also includes strong participation from industry with an Industrial Advisory Board composed of key experts from University as well as Automotive, Space and Microprocessor industries and experts in Real-Time Operating Systems, Compilers and Software tools for critical real-time embedded systems.

Technical Vision

Main Objectives

Technical Approach

Key Issues

Expected Impact