WHICH CONCURRENT ERROR DETECTION SCHEME TO CHOOSE?
ITC '00 Proceedings of the 2000 IEEE International Test Conference
Analysis of Error Recovery Schemes for Networks on Chips
IEEE Design & Test
Exploring Fault-Tolerant Network-on-Chip Architectures
DSN '06 Proceedings of the International Conference on Dependable Systems and Networks
Low-cost router microarchitecture for on-chip networks
Proceedings of the 42nd Annual IEEE/ACM International Symposium on Microarchitecture
Exploiting inherent information redundancy to manage transient errors in NoC routing arbitration
NOCS '11 Proceedings of the Fifth ACM/IEEE International Symposium on Networks-on-Chip
Hi-index | 0.00 |
Reliable operation that can be checked on-line is of paramount importance to current and future systems-on-chips that are implemented in very deep submicron technologies. In such systems, the communication among architectural modules is handled by a modular network-on-chip infrastructure that should be sufficiently protected from transient faults that may affect its correct operation. The error protection mechanism should cover all fault scenarios and incur the minimum area/energy/delay overhead. In this paper, we propose such an on-line checking mechanism for the switch allocator of the router that detects every possible single transient or permanent fault in the arbiters and handles it appropriately, thus preserving the reliable operation of the switch.