On-Demand Redundancy Grouping: Selectable Soft-Error Tolerance for a Multicore Cluster

TL;DR

This paper presents a flexible architectural scheme called On-Demand Redundancy Grouping (ODRG) for multicore clusters, enabling run-time soft-error tolerance with minimal overhead and performance benefits.

Contribution

It introduces a novel run-time configurable fault-tolerance scheme for multicore clusters, allowing dynamic switching between fault-tolerant and high-performance modes.

Findings

Less than 11% core area overhead for redundancy grouping

Negligible timing increase with the ODRG scheme

2.5× faster fault recovery compared to state-of-the-art

Abstract

With the shrinking of technology nodes and the use of parallel processor clusters in hostile and critical environments, such as space, run-time faults caused by radiation are a serious cross-cutting concern, also impacting architectural design. This paper introduces an architectural approach to run-time configurable soft-error tolerance at the core level, augmenting a six-core open-source RISC-V cluster with a novel On-Demand Redundancy Grouping (ODRG) scheme. ODRG allows the cluster to operate either as two fault-tolerant cores, or six individual cores for high-performance, with limited overhead to switch between these modes during run-time. The ODRG unit adds less than 11% of a core's area for a three-core group, or a total of 1% of the cluster area, and shows negligible timing increase, which compares favorably to a commercial state-of-the-art implementation, and is 2.5$\times$ faster in fault recovery re-synchronization. Furthermore, when redundancy is not necessary, the ODRG approach allows the redundant cores to be used for independent computation, allowing up to 2.96$\times$ increase in performance for selected applications.