TY - GEN
T1 - Distributed control framework for mapreduce cloud on cloud computing
AU - Huang, Tzu Chi
AU - Chu, Kuo Chih
AU - Huang, Guo Hao
AU - Shen, Yan Chen
AU - Shieh, Ce Kuen
N1 - Funding Information:
We thank the Taiwan Ministry of Science and Technology for the supports of this project under grant number MOST 106-2221-E-262-004. We thank Lunghwa University of Science and Technology for kindly providing us with devices. We further offer our thanks to the reviewers for their comments.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/7/6
Y1 - 2018/7/6
N2 - A MapReduce cloud becomes a key to the success of cloud computing today. However, a MapReduce cloud uses a single Master node as the brain to manage tasks distributed over Slave nodes for controlling the entire progress of the application execution. Accordingly, a MapReduce cloud easily overloads the Master node with reports sent from Slave nodes at run time to harm performance. Besides, a MapReduce cloud makes the Master node a single failure point to suspend the application execution when the Master node cannot work. A MapReduce cloud can use the Distributed Control Framework (DCF) proposed in this paper to improve both performance and fault tolerance, because DCF shifts most works of a Master node to a DCF Master Agent coexisting in each Slave node and allows Slave nodes to join or leave a cloud at run time without interrupting the application execution. According to observations on experiments with various applications in this paper, a MapReduce cloud can use DCF to have better performance and fault tolerance in comparison to a native MapReduce cloud.
AB - A MapReduce cloud becomes a key to the success of cloud computing today. However, a MapReduce cloud uses a single Master node as the brain to manage tasks distributed over Slave nodes for controlling the entire progress of the application execution. Accordingly, a MapReduce cloud easily overloads the Master node with reports sent from Slave nodes at run time to harm performance. Besides, a MapReduce cloud makes the Master node a single failure point to suspend the application execution when the Master node cannot work. A MapReduce cloud can use the Distributed Control Framework (DCF) proposed in this paper to improve both performance and fault tolerance, because DCF shifts most works of a Master node to a DCF Master Agent coexisting in each Slave node and allows Slave nodes to join or leave a cloud at run time without interrupting the application execution. According to observations on experiments with various applications in this paper, a MapReduce cloud can use DCF to have better performance and fault tolerance in comparison to a native MapReduce cloud.
UR - http://www.scopus.com/inward/record.url?scp=85050653108&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85050653108&partnerID=8YFLogxK
U2 - 10.1109/NOMS.2018.8406180
DO - 10.1109/NOMS.2018.8406180
M3 - Conference contribution
AN - SCOPUS:85050653108
T3 - IEEE/IFIP Network Operations and Management Symposium: Cognitive Management in a Cyber World, NOMS 2018
SP - 1
EP - 4
BT - IEEE/IFIP Network Operations and Management Symposium
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 IEEE/IFIP Network Operations and Management Symposium, NOMS 2018
Y2 - 23 April 2018 through 27 April 2018
ER -