On Structuring Holistic Fault Tolerance

Gensh, R; Romanovsky, A; Yakovlev, A

doi:10.1145/2889443.2889458

On Structuring Holistic Fault Tolerance

Lookup NU author(s): Rem Gensh, Professor Alexander Romanovsky ORCiD, Professor Alex Yakovlev

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

Computer systems are developed taking into account that they should be easily maintained in the future. It is one of the main requirements for the sound architectural design. The existing approaches to introducing fault tolerance rely on recursive system structuring out of functional components - this typically results in non-optimal fault tolerance. The paper proposes a vision of structuring complex many-core systems by introducing a special component supporting system-wide fault tolerance coordination. The component acts as a central module making decisions about fault tolerance strategies to be implemented by individual system components depending on the performance and energy requirements specified as system operating modes.