作者: Sung Soo Kim , Joong Moo Park , Jang Kyung Kim , Kie Jin Park , Sang Hyun Kim
DOI:
关键词: Software availability 、 Failure rate 、 Software system 、 Engineering 、 Software rejuvenation 、 Software 、 Server 、 Cluster (spacecraft) 、 Operating system 、 Embedded system 、 Computer cluster
摘要: The invention relates to a method and apparatus for improving software availability of cluster computer system via rejuvenation technique, in which program is temporarily stopped at an adequate time point that manager constituted by several servers can expect, then restarted. In the invention, both aspects hardware are considered, proactive fault-tolerance technique utilized improved through determination optimal period according unstable rate failure so features high-available be ensured efficient cost.