Optimal Checkpoint Times for Grouped Redundant Arrays Inexpensive Computer Nodes (GRAICN)  
Author Kenichiro Naruse

 
Co-Author(s) Toshio Nakagawa

 
Abstract Tasks with random processing times are executed successively. Two types of checkpoints can be placed at the end of tasks. We suppose several databasenodes and one management-node. Tasks are executed by the database-nodes and are watched by management-node. Two types of checkpoints are called Journal Checkpoint (JC) and Flush Checkpoint (FC). JC needs a small time compared with FC to store checkpoint data, and FC needs a long time compared with JC to store checkpoint data. To get high reliability, we adopt RAID 5 architecture to grouped multi-database-node system. The problem is that in what places we set suitable checkpoints, and how many number of database-nodes and groups we should prepare. We derive how many number of JC checkpoints between FC to minimize the execute time, and how many number of database-nodes to minimize it. The mean execution times per one task are obtained and optimal numbers that minimize them are derived analytically and numerically.

 
Keywords Reliability, Checkpoint times, Big data, Database system, RAID systems, High speed processing
   
    Article #:  2132
 
Proceedings of the 21st ISSAT International Conference on Reliability and Quality in Design
August 6-8, 2015 - Philadelphia, Pennsylvia, U.S.A.