#	Pagina
attuale pagina	/open-h2020/projects/194260/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 2 - IMMORTAL (Integrated Modelling, Fault Management, Verification and Reliable Design Environment for Cyber-Physical Systems)

Teaser

Summary

In IMMORTAL a consortium of leading European academic and industrial players aim at combining their expertise in developing an integrated, cross-layer modelling based tool framework for fault management, verification and reliable design of dependable cyber-physical systems.
Recently, the world has seen emerging Cyber-Physical System (CPS) modelling frameworks addressing various design aspects such as control, security, verification and validation. However, there have been no considerations for reliability and automated debug (i.e. design error localisation and correction) aspects. The main aim of IMMORTAL is to fill this gap by introducing reliable design and automated system debug into CPS modelling. To reach this aim, the project will develop a cross-layer CPS model spanning analogue mixed-signal circuits, hardware architecture, firmware, operating system and application layers. In addition, a holistic fault model for representing fundamentally different error sources in CPSs (design bugs, wear-out and environmental effects) in a uniform manner will be proposed. Moreover, IMMORTAL plans to develop a fault management infrastructure on top of the reliable design framework that would allow ultra-fast fault detection, isolation and recovery in the emerging many-core based CPS architectures that are expected to be increasingly adopted in the coming years.
As a result, the project will enable the development of dependable CPSs with improved reliability and extended effective life-time, which is a particular concern in emerging nanoelectronics technologies that are becoming increasingly vulnerable to disturbances, ageing and process variations. The tool framework to be developed within IMMORTAL will be evaluated on a clearly specified real-world use-case of a satellite on-board-computer. However, since the results are more general and applicable to many application domains, including avionics, automotive and telecommunication, demonstration of the framework tools will be applied to CPS examples from other domains as well.

Work performed

The work has so far led to several major milestones by successfully developing the following: a holistic CPS modelling framework; reliability metrics and fault models; verification, debugging and testing methods; dynamic, semiformal and formal reliability analysis methods and online fault detection mechanisms.
Activities during the second period included organization of a two-day Dissemination Workshop RENSâ€™16 collocated with IFIP/IEEE VLSI-SoC and a Special Session organised on â€œDesigning Reliable Cyber-Physical Systemsâ€ at Forum on specification & Design Languages (FDL\'16).

In addition, a patent has been submitted, more than 30 scientific articles have been published and several presentations, including as much as four conference keynotes and many invited talks and panel performances on the projectâ€™s topic have taken place.

Final results

IMMORTAL goes beyond state of the art in CPS development in three areas: reliable design, fault management and automated debug.
IMMORTAL addresses the problem of analysing and verifying reliability aspects of the hardware components of CPS. The goal is to develop complete and automated methods for reliability analysis. In addition, high-level reliability models will be developed that will combine the per-component analyses into a system-wide reliability characterisation.
In fault management, the goal is to achieve early fault detection and fast recovery by implementing a cross-layer fault management approach. This enables a graceful degradation environment for the CPS, where the systemsâ€™ tolerance to faults and life-time is improved and costs for maintenance are significantly reduced.
Concerning automated debug, IMMORTAL will develop verification engines for CPS with a goal to extend them by automated design error localisation and correction capabilities. Currently, solutions for automated debug in CPSs are missing. Yet, it has been shown by numerous studies that error localisation and correction in digital computing systems consume a major portion of the overall development effort. In CPSs, this problem is going to be even more severe due to the underlying complexity and heterogeneity.
IMMORTAL innovations are expected to lead to the following improvements.
1) Minimisation of the verification effort in CPSs by a factor of 2 by enabling automated debug (error localisation and correction) in such systems. Methods that rely on lightweight models for CPS will be developed therefore, improving the scalability.
2) Speeding up fault detection, isolation and recovery in CPSs by a factor of 4 by implementing a cross-layer approach, a holistic fault model and a new fault management architecture.
3) Graceful degradation: by resumption of correct operation with up to 15% of CPS network resources failed. To be achieved by development of network reconfiguration, fault localisation and resource isolation schemes for CPSs based on many-core networks.
4) Up to 40% reduction in the effort designers put in reliability related tasks by developing an automated and complete sign-off tool.
5) Up to 10% savings in the total area as well as power consumption achieved by optimising hardware protection logic overhead.

These improvements would lead to the following overall expected impacts to future CPSs:
â€¢ 30% reduction of development time
â€¢ 40% reduction in maintenance costs
This will translate to cheaper yet dependable CPSs for the society.