Modeling and Analysis of Grid Service Reliability Considering Fault Recovery

Suchang Guo, Hong-Zhong Huang, Yu Liu
2011 New generation computing  
The extreme complexity of grid system makes it extremely difficult to achieve high service reliability, and this situation is aggravated by the fact that many grid services need to perform time-consuming tasks that may require several days or even months of computation. To improve grid service reliability, this paper studies a fault recovery technique in grid systems and conducts in-depth research on grid reliability modeling and analysis with fault recovery. Grid failures considered in this
more » ... er are classified into two categories: unrecoverable failures and recoverable failures. Software reliability is taken into account as well. To make fault recovery more practical, certain constraints on fault recovery are introduced and grid service reliability models under these practical constraints are developed. Numerical examples are presented, and based on the results obtained, the impact of fault recovery as well as that of practical constraints on grid service reliability is discussed.
doi:10.1007/s00354-009-0114-8 fatcat:xp3knypub5g2nbsp5dq7xl7qea