Error-control coding for computer systems
Error-control coding for computer systems
Design & analysis of fault tolerant digital systems
Design & analysis of fault tolerant digital systems
Fault-Tolerant Design Strategies for High Reliability and Safety
IEEE Transactions on Computers
Duplex and Triplex Memory: Which Is More Reliable?
Duplex and Triplex Memory: Which Is More Reliable?
SEU tolerant device, circuit and processor design
Proceedings of the 42nd annual Design Automation Conference
Hi-index | 14.98 |
A large number of choices exist when designing a reliable memory system. The choices range from simple replication to complex error control codes (ECC). An intermediate solution is to use combination of replication and simple ECC. Such a system consists of multiple memory modules, data stored in each module being encoded using an ECC. This paper compares reliability of memory systems formed using simple triplication (without ECC) with memory systems formed by duplicating memory modules that use ECC. It is shown that reliability achieved by duplication of memory modules using codes capable of only error detection or only single error correction (SEC), is always worse than simple triplication. However, it is also shown that duplication of memory modules, with codes capable of single error correction and double error detection (SEC-DED), can achieve better reliability than simple triplication when bit error probability is small.