TY - GEN
T1 - Load balancing in the cloud using specialization
AU - Hammoudi, Sarra
AU - Benaouda, Abdelhafid
AU - Harous, Saad
AU - Aliouat, Zibouda
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/12/7
Y1 - 2016/12/7
N2 - On the Internet of the future, billions of physical objects will collect and exchange data that will be stored in the Cloud Computing. A bad storage structure of big data has a negative influence on the latency time during the client's requests. To solve this problem, we have proposed an Infrastructure as a Service (IaaS) for the input and the output of different files on servers' clusters. We have proposed a specialization based architecture and used a multi-agents system paradigm. Some proposed agents collaborate to realize the load balancing, while others work in a parallel way to achieve their tasks rapidly and minimize the latency time. We implemented our solution using the JADE platform. We measured the execution time and the response time for each request type (input/output) and for the file type (text, image, and video). By comparing the execution time and the response time using our architecture and the classic one, our architecture has given better results.
AB - On the Internet of the future, billions of physical objects will collect and exchange data that will be stored in the Cloud Computing. A bad storage structure of big data has a negative influence on the latency time during the client's requests. To solve this problem, we have proposed an Infrastructure as a Service (IaaS) for the input and the output of different files on servers' clusters. We have proposed a specialization based architecture and used a multi-agents system paradigm. Some proposed agents collaborate to realize the load balancing, while others work in a parallel way to achieve their tasks rapidly and minimize the latency time. We implemented our solution using the JADE platform. We measured the execution time and the response time for each request type (input/output) and for the file type (text, image, and video). By comparing the execution time and the response time using our architecture and the classic one, our architecture has given better results.
KW - Cloud computing
KW - big data
KW - dispatcher
KW - load balancing
KW - multi-agent systems
KW - servers clusters
KW - wireless sensor network
UR - http://www.scopus.com/inward/record.url?scp=85010380073&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85010380073&partnerID=8YFLogxK
U2 - 10.1109/UEMCON.2016.7777853
DO - 10.1109/UEMCON.2016.7777853
M3 - Conference contribution
AN - SCOPUS:85010380073
T3 - 2016 IEEE 7th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2016
BT - 2016 IEEE 7th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2016
A2 - Saha, Himadri Nath
A2 - Chakrabarti, Satyajit
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 7th IEEE Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2016
Y2 - 20 October 2016 through 22 October 2016
ER -