PAReS: A Proactive and Adaptive Redundant System for MapReduce
JIA-CHUN LIN；FANG-YIE LEU；YING-PING CHEN
MapReduce ； redundant system ； service downtime ； takeover ； proactive synchronization and replication ； mutual monitoring ； adaptive warm-up mechanism
Journal of Information Science and Engineering
|Volume or Term/Year and Month of Publication||
31卷5期（2015 / 09 / 01）
1797 - 1815
Recently, MapReduce has been a key and popular technology for tackling data-intensive applications. But its two master servers in current MapReduce implementations have a single-failure problem, which may interrupt MapReduce operations and filesystem services. In this paper, we propose a hybrid takeover scheme, called the Proactive and Adaptive Redundant System (PAReS for short), which employs three service-quality improvement mechanisms, including a proactive synchronization and replication method, a mutual monitoring algorithm, and an adaptive warm-up mechanism, to mitigate the above problems. The extensive experiments show that PAReS enhances service quality at acceptable energy consumption level and synchronization cost as compared with four stateof- the-art schemes.