Building a fault tolerant application using the GASPI communication layer

TL;DR

This paper demonstrates how to build fault-tolerant applications using the GASPI communication layer, focusing on low-cost failure detection and recovery mechanisms suitable for exascale computing.

Contribution

It introduces a fault-tolerant application framework based on GASPI, extending checkpointing with efficient failure detection and recovery, and analyzes its overhead and scalability.

Findings

Failure detection causes no overhead in failure-free runs

Recovery overhead is acceptable and scales well

The approach effectively handles process failures in exascale environments

Abstract

It is commonly agreed that highly parallel software on Exascale computers will suffer from many more runtime failures due to the decreasing trend in the mean time to failures (MTTF). Therefore, it is not surprising that a lot of research is going on in the area of fault tolerance and fault mitigation. Applications should survive a failure and/or be able to recover with minimal cost. MPI is not yet very mature in handling failures, the User-Level Failure Mitigation (ULFM) proposal being currently the most promising approach is still in its prototype phase. In our work we use GASPI, which is a relatively new communication library based on the PGAS model. It provides the missing features to allow the design of fault-tolerant applications. Instead of introducing algorithm-based fault tolerance in its true sense, we demonstrate how we can build on (existing) clever checkpointing and extend applications to allow integrate a low cost fault detection mechanism and, if necessary, recover the application on the fly. The aspects of process management, the restoration of groups and the recovery mechanism is presented in detail. We use a sparse matrix vector multiplication based application to perform the analysis of the overhead introduced by such modifications. Our fault detection mechanism causes no overhead in failure-free cases, whereas in case of failure(s), the failure detection and recovery cost is of reasonably acceptable order and shows good scalability.