A Cost Effective Spacial Redundancy with Data-Path Partitioning

Abstract

In order to maintain the high reliability of a computer system, it is necessary to detect the failure leading to a fault. In general, fault can be detected by exploiting time redundancy or spatial redundancy. However, it negatively affects on either hardware cost or processor performance. To solve the cost-performance issue, in this paper, we propose a concept of cost-effective approach to achieve spatial redundancy for dependable processors. In addition, we perform a primly evaluation for the impact of our method on processor performance