Application-level checkpointing for shared memory programs

Abstract

Trends in high-performance computing are making it necessary for long-running applications to tolerate hardware faults. The most commonly used approach is checkpoint and restart (CPR) -the state of the computation is saved periodically on disk, and when a failure occurs, the computation is restarted from the last saved state. At present, it is the responsibility of the programmer to instrument applications for CPR.Our group is investigating the use of compiler technology to instrument codes to make them self-checkpointing and self-restarting, thereby providing an automatic solution to the problem of making long-running scientific applications resilient to hardware faults. Our previous work focused on message-passing programs.In this paper, we describe such a system for sharedmemory programs running on symmetric multiprocessors. This system has two components: (i) a pre-compiler for source-to-source modification of applications, and (ii) a runtime system that implements a protocol for coordinating CPR among the threads of the parallel application. For * the sake of concreteness, we focus on a non-trivial subset of OpenMP that includes barriers and locks.One of the advantages of this approach is that the ability to tolerate faults becomes embedded within the application itself, so applications become self-checkpointing and self-restarting on any platform. We demonstrate this by showing that our transformed benchmarks can checkpoint and restart on three different platforms (Windows/x86, Linux/x86, and Tru64/Alpha). Our experiments show that the overhead introduced by this approach is usually quite small; they also suggest ways in which the current implementation can be tuned to reduced overheads further.

A technique for non-invasive application-level checkpointing

“…Bronevetsky et al [6][7][8] have proposed a preprocessor-based approach for ALC. Their work is relevant for both shared memory and distributed memory architectures and their approach consists of two components: a preprocessor, and a checkpointing library.…”

“…During the domain analysis phase of developing the DSL, a survey of technical literature and existing implementations [6,7,14,15,20] was done to obtain an overview of the terminologies and concepts related to the ALC-domain. Commonly used terms and their relationships were used to develop the domain lexicon.…”

“…Together, these steps are referred to as Checkpointing and Restart (CaR) through the rest of this paper. The main types of checkpointing techniques, depending upon the level of transparency, are: hardware-level [3], system-level [4], userlevel [5], application-level [6][7][8][9], and hybrid approaches [10].…”

Lightweight checkpointing for concurrent ML

“…Existing checkpoint approaches can be classified into four broad categories: (a) schemes that require applications to provide their own specialized checkpoint and recovery mechanisms (Bronevetsky et al 2003;Bronevetsky et al 2004); (b) schemes in which the compiler determines where checkpoints can be safely inserted (Beck et al 1994); (c) techniques that require operating system or hardware monitoring of thread state (Li et al 1990;Hulse 1995;Chen et al 1997); and (d) library implementations that capture and restore state (Dieter & Lumpp 1999). Checkpointing functionality provided by an application or a library relies on the programmer to define meaningful checkpoints.…”

Coarse grain computation-communication overlap for efficient application-level checkpointing for GPUs

“…Cornell Checkpointing Compiler (C 3 ) is an application level checkpointing implementation for multi-thread applications in shared memory [12]. The checkpoint is done in three steps: each thread (1) calls a barrier, (2) saves its private state, and (3) calls a second barrier.…”