Method for improving performance in online routing of reconfigurable nano architectures

Niknahad, Mahtab; Hübner, Michael; Becker, Jürgen

doi:10.1109/soccon.2009.5398093

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However none of these methods use the idea of improving the system's reliability situation during runtime and therefore decreasing the different costs of fault tolerance. In the [15][16] [17] we have introduced our ideas about dynamic and static fault tolerance in nano scale architectures. In [15] we have used defectmaps to recognize the faulty places and in [17] we have avoided the faulty places using rerouting .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reliability Analysis and Improvement in Nano Scale Design

Niknahad

Huebner

Becker

2010

2010 IEEE Computer Society Annual Symposium on VLSI

Self Cite

View full text Add to dashboard Cite

According to the shrinking feature size of the VLSI circuits it is expected that nano scale devices and interconnections will introduce unprecedented level of defects and architectural designs need to settle with the uncertainty result at such scales. Several approaches for implementing the fault tolerance systems are already investigated. Most of these methods are applicable also in the case of high fault rates. Most protection methods are based on different redundancy methods which add extra detection and correction features to the design. We strongly believe that in future architectures it become more important assessing the fault tolerance techniques. Having an estimation of system fault tolerance can ensure critical applications working properly. In this work we propose a new method which checks reconfigurable architectures and during runtime finds violent spots in the design for probable transient and permanent failures. This approach is adjustable to either current FPGAs or future nano-architectures which are based on reconfigurability. We define a fault detection model for probable errors which uses an efficient algorithm that proves the fault tolerance in the reconfigurable architecture and computes a reliability factor for the architecture. This helps avoiding using the critical parts by future usages. Our method is applicable to different levels of granularity, such as gate level, logic block level, logic function level, unit level, etc. It is efficient and fast and can be simply integrated into the design flow.

show abstract

Section: Introductionmentioning

confidence: 99%

“…In the [15][16] [17] we have introduced our ideas about dynamic and static fault tolerance in nano scale architectures. In [15] we have used defectmaps to recognize the faulty places and in [17] we have avoided the faulty places using rerouting .…”

Section: Introductionmentioning

confidence: 99%