Spare sharing network enhancement for scalable systems

Khaleghi, Soroush; Rao, Wenjing

doi:10.1109/dft.2013.6653614

Cited by 4 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For f faults occurring sequentially on the |N U | functional units, there is a total number of |N U | f equally possible fault occurrence patterns, including repeated fault occurrence onto the same functional unit (i.e., the spares that take over its job). For any of the |N U | f fault occurrence patterns, it is defined as repairable if at least one successful repair sequence exists for it [15] [23].…”

Section: Repairabilitymentioning

confidence: 99%

“…According to the repairability model, an immediate replacement failure happens when a faulty functional unit cannot be replaced by any spares, i.e., the fan-out degree of the faulty functional unit is equal to zero. Consequently, the best way to construct a network would be to distribute all edges evenly in a regular manner among functional units and spare units [15] [23]. A high-order ring structure, for which all the functional units have a same fan-out degree, and so do all the spare units, can offer such an "optimal" repairability for a given number of edges.…”

Section: Network Construction Enhancementmentioning

confidence: 99%

“…A high-order ring structure, for which all the functional units have a same fan-out degree, and so do all the spare units, can offer such an "optimal" repairability for a given number of edges. However, due to several interconnection constraints such as wiring area, routing delays, and power consumption, as well as the functionality match constraint, it is not always possible or desirable to have a perfect ring structure network [15] [23]. In this work, we focus on the general framework of improving the repairability of any arbitrary network, with sparse sharing of spare units.…”

Section: Network Construction Enhancementmentioning

confidence: 99%

See 2 more Smart Citations

Repairability Enhancement of Scalable Systems with Locally Shared Spares

Khaleghi,

Rao

2018

Preprint

Self Cite

View full text Add to dashboard Cite

Future systems based on nano-scale devices will provide great potentials for scaling up in system complexity, yet they will be highly susceptible to operational faults. While spare units can be generally used to enhance reliability, they must be shared in a limited way among functional units to ensure low-cost overheads when systems scale up. Furthermore, the efficiency of achieving reliability using spare units heavily depends on the replacement mechanisms of such spares. While global and chained replacement approaches can take advantage of the entire replacement capabilities in the network, they usually impose some sort of disturbance to all the functional units in the system during the repair process, thus are dreadfully expensive in terms of performance overhead for systems with high fault rates. In this paper, we focus on a low-cost, fast, "immediate" replacement mechanism that can be implemented locally with minimum disturbance to the system. The proposed schemes aim for maintaining the system with high fault rates in such a lowcost, fast repairable status for many faults before invoking the more expensive, yet optimal, approaches. First, we propose an online repair algorithm: as faults occur during the run-time of the system, the proposed algorithm makes a choice of a spare unit (among several candidates), such that the overall impact on system repairability in the future is minimized. Second, we propose a network enhancement approach, which identifies and connects the vulnerable units to the exploitable spares, thus strengthening the entire system at a low cost.

show abstract

Section: Repairabilitymentioning

confidence: 99%

Section: Network Construction Enhancementmentioning

confidence: 99%

Section: Network Construction Enhancementmentioning

confidence: 99%

See 1 more Smart Citation

Repairability Enhancement of Scalable Systems with Locally Shared Spares

Khaleghi,

Rao

2018

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…This shows that the choice of the healthy PE that shall replace a faulty one, and the resulting topology have a significant impact on the overall performance. Khaleghi and Rao [2013] addressed the problem of how to efficiently integrate spare cores in the original NoC by implementing a spare sharing network containing a small number of connections to the spare cores. Their integration process also considered router faults, so that router faults do not prevent the utilization of healthy PEs.…”

Section: Spare Coresmentioning

confidence: 99%

A Survey on Design Approaches to Circumvent Permanent Faults in Networks-on-Chip

2016

View full text Add to dashboard Cite

Increasing fault rates in current and future technology nodes coupled with on-chip components in the hundreds calls for robust and fault-tolerant Network-on-Chip (NoC) designs. Given the central role of NoCs in today's many-core chips, permanent faults impeding their original functionality may significantly influence performance, energy consumption, and correct operation of the entire system. As a result, fault-tolerant NoC design gained much attention in recent years. In this article, we review the vast research efforts regarding a NoC's components, namely, topology, routing algorithm, router microarchitecture, as well as system-level approaches combined with reconfiguration; discuss the proposed architectures; and identify outstanding research questions.

show abstract

A local reconfiguration based scalable fault tolerant many-processor array

Banerjee

Rao

2017

2017 22nd Asia and South Pacific Design Automation Conference (ASP-DAC)

View full text Add to dashboard Cite

Spare sharing network enhancement for scalable systems

Cited by 4 publications

References 12 publications

Repairability Enhancement of Scalable Systems with Locally Shared Spares

Repairability Enhancement of Scalable Systems with Locally Shared Spares

A Survey on Design Approaches to Circumvent Permanent Faults in Networks-on-Chip

A local reconfiguration based scalable fault tolerant many-processor array

Contact Info

Product

Resources

About