Check-pointing approach for fault tolerance in OpenSHMEM

Abstract

Fault tolerance for long running applications is critical to guard against failure of either compute resources or a network. Accomplishing this task in software is non-trivial and there is an added level of complexity for implementing a working model for a one-sided communications library like OpenSHMEM, since there is no matching communication call at the target processing element (PE). In this paper we explore a fault tolerance scheme based on check-point and restart, that caters to the one-sided nature of PGAS programming model while leveraging features very specific to OpenSHMEM. Through a working implementation with the 1-D Jacobi code, we show that the approach is scalable and provides considerable computational resource saving.