YLYawei Li

Yawei Li

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yawei Li has a total of 1 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Analytical study of migration-enhanced fault tolerance for long-running applications in IFR systems

    by Yawei Li, Zhiling Lan

    Computer systems with increasing failure rate (IFR) are common in practice. For such systems, the literature indicates that aperiodic checkpointing can provide better performance than periodic checkpointing due to its adaptability to the failure process. However, for long-running applications, aperiodic checkpointing suffers from substantial operational overhead due to frequent checkpointing operations as the application proceeds. To address this problem, in this paper, we propose to incorporate just-in-time process migration in addition to aperiodic checkpointing for applications running in an IFR system. The goal is to reduce application execution time in the presence of failures. In particular, we present an analytical study of this migration-enhanced fault tolerance scheme (denoted as migCP) by deriving application completion time by using migCP and further determining the optimal migration locations. We demonstrate, through analytical modelling and empirical studies, that migCP outperforms aperiodic checkpointing under a variety of system parameters.

  • Current Research and Practice in Proactive Fault Management

    by Y. Li, Z. Lan

    Unlike rollback-recovery, proactive fault management takes preventive actions before the occurrence of failures. In this survey paper, we classify the current research of proactive fault management into two broad categories: failure analysis and prediction, and proactive techniques. Analytical methods have been widely used to analyse and forecast contiguous values, while data mining or machine learning methods are mostly suited to categorical data. Various proactive fault management systems have been recently developed, each of them exploring different proactive techniques to achieve its specific design goal. Our investigation shows that research should be conducted in the context of high performance computing to enable efficient proactive fault management for the emerging large-scale supercomputers.