The Ravenscar Tasking Profile for High Integrity Real-Time Programs
Ada-Europe '98 Proceedings of the 1998 Ada-Europe International Conference on Reliable Software Technologies
Preemptive Scheduling of Multi-criticality Systems with Varying Degrees of Execution Time Assurance
RTSS '07 Proceedings of the 28th IEEE International Real-Time Systems Symposium
The worst-case execution-time problem—overview of methods and survey of tools
ACM Transactions on Embedded Computing Systems (TECS)
Towards a ravenscar extension for multi-processor systems
ACM SIGAda Ada Letters
Partitioned Embedded Architecture Based on Hypervisor: The XtratuM Approach
EDCC '10 Proceedings of the 2010 European Dependable Computing Conference
Mixed-Criticality Real-Time Scheduling for Multicore Systems
CIT '10 Proceedings of the 2010 10th IEEE International Conference on Computer and Information Technology
A survey of hard real-time scheduling for multiprocessor systems
ACM Computing Surveys (CSUR)
Design and implementation of a Ravenscar extension for multiprocessors
Ada-Europe'11 Proceedings of the 16th Ada-Europe international conference on Reliable software technologies
ORK+/XtratuM: an open partitioning platform for Ada
Ada-Europe'11 Proceedings of the 16th Ada-Europe international conference on Reliable software technologies
Implementing mixed criticality systems in Ada
Ada-Europe'11 Proceedings of the 16th Ada-Europe international conference on Reliable software technologies
Hierarchical scheduling with ada 2005
Ada-Europe'06 Proceedings of the 11th Ada-Europe international conference on Reliable Software Technologies
Cache-Aware development of high-integrity systems
Ada-Europe'10 Proceedings of the 15th Ada-Europe international conference on Reliable Software Technologies
Hi-index | 0.00 |
Partitioning is a common approach to developing mixed-criticality systems, where partitions are isolated from each other both in the temporal and the spatial domain in order to prevent low-criticality subsystems from compromising other subsystems with high level of criticality in case of misbehaviour. The advent of many-core processors, on the other hand, opens the way to highly parallel systems in which all partitions can be allocated to dedicated processor cores. This trend will simplify processor scheduling, although other issues such as mutual interference in the temporal domain may arise as a consequence of memory and device sharing. The paper describes an architecture for multi-core partitioned systems including critical subsystems built with the Ada Ravenscar profile. Some implementation issues are discussed, and experience on implementing the ORK kernel on the XtratuM partitioning hypervisor is presented.