A checkpointing‐enabled and resource‐aware Java Virtual Machine for efficient and robust e‐Science applications in grid environments

Abstract

Object-oriented programming languages presently are the dominant paradigm of application development (e. g., Java,. NET). Lately, increasingly more Java applications have long (or very long) execution times and manipulate large amounts of data/information, gaining relevance in fields related with e-Science (with Grid and Cloud computing). Significant examples include Chemistry, Computational Biology and Bio-informatics, with many available Java-based APIs (e. g., Neobio).Often, when the execution of such an application is terminated abruptly because of a failure (regardless of the cause being a hardware of software fault, lack of available resources, etc.), all of its work already performed is simply lost, and when the application is later re-initiated, it has to restart all its work from scratch, wasting resources and time, while also being prone to another failure and may delay its completion with no deadline guarantees.Our proposed solution to address these issues is through incorporating mechanisms for checkpointing and migration in a JVM. These make applications more robust and flexible by being able to move to other nodes, without any intervention from the programmer. This article provides a solution to Java applications with long execution times, by extending a JVM (Jikes research virtual machine) with such mechanisms.

Evaluation of an adaptive framework for resilient Monte Carlo executions

“…Among them, the implementation of a fault tolerant Java Virtual Machine that supports checkpointing is proposed in [14], and the robustness of the critical paths of workflows is ensured in [15]. In both cases, an issue of the work is to find a balance between resource consumption and desired reliability.…”

A progress and profile-driven cloud-VM for resource-efficiency and fairness in e-science environments

“…In order to comply with this requisite, each instance of ARA-JVM is enhanced with services that allow for: i) monitor the application progress, ii) account resource consumption, iii) reconfigure internal parameters and/or mechanisms, and iv) checkpoint, restore and migrate the whole application. In [12] we focus on the last point which regards checkpointing and restore. Our current design and implementation effort mainly concerns the progress monitoring module and the resource scheduler.…”

Ditto – Deterministic Execution Replayability-as-a-Service for Java VM on Multiprocessors

“…Furthermore, the replay can only be necessary to be done from a certain point in time because the fault is known to occur only at the end of execution. Ditto uses a lightweight checkpointing mechanism [17] to offer two new replay services: (i) replay to most recent point before fault; (ii) replay to any instant M in execution. Checkpoint is done recording each thread stack and reachable objects.…”

“…In this case, the total recording space is N * sizeof (checkpoint) + N * sizeof (truncatedLog), where N is the number of times a checkpoint is done. In this case there is a trade-off between overhead in execution time and granularity in available replay start times [17]. Even so, the total recording space is bounded to be smaller than 2 * N * sizeof (checkpoint).…”