A robust multiplexer-based FPGA inspired by biological systems
Journal of Systems Architecture: the EUROMICRO Journal - Special issue: dependable parallel computer systems
Efficiently supporting fault-tolerance in FPGAs
FPGA '98 Proceedings of the 1998 ACM/SIGDA sixth international symposium on Field programmable gate arrays
Methodologies for Tolerating Cell and Interconnect Faults in FPGAs
IEEE Transactions on Computers
Tolerating operational faults in cluster-based FPGAs
FPGA '00 Proceedings of the 2000 ACM/SIGDA eighth international symposium on Field programmable gate arrays
Efficient incremental rerouting for fault reconfiguration in field programmable gate arrays
ICCAD '99 Proceedings of the 1999 IEEE/ACM international conference on Computer-aided design
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
On Routability for FPGAs under Faulty Conditions
IEEE Transactions on Computers
FPL '97 Proceedings of the 7th International Workshop on Field-Programmable Logic and Applications
Dynamic Fault Tolerance in FPGAs via Partial Reconfiguration
FCCM '00 Proceedings of the 2000 IEEE Symposium on Field-Programmable Custom Computing Machines
Using Roving STARs for On-Line Testing and Diagnosis of FPGAs in Fault-Tolerant Applications
ITC '99 Proceedings of the 1999 IEEE International Test Conference
Hi-index | 0.00 |
In this paper we analyze the performance penalty of a fault-tolerant (FT) adaptive computing system (ACS) that implements the roving Self Testing AReas (STARs) approach for on-line testing and fault tolerance for FPGAs[1,5]. For most benchmarks, the presence of the STARs increases the critical path delay by 4.6% to 22%, and preallocating spare cells for fault tolerance causes an additional increase of up to 37%. We also present a procedure for estimating the worst case performance penalty caused by an incremental change of an already placed and routed FPGA. This estimate can be used to guide the selection of fault-tolerant reconfigurations to minimize their impact on the system timing. Our results show that the estimate is within 10% of the real delay values.