DocumentCode :
3030897
Title :
Robustness Analysis of Efficient & Scalable Priority Based Jingle-Mingle Model Using Mosix
Author :
Gupta, Ankit ; Agarwal, Rahul ; Vashishtha, Rohit ; Mishra, Shakti ; Kushwaha, D.S.
Author_Institution :
CSED, MNNIT, Allahabad, India
fYear :
2009
fDate :
28-29 Dec. 2009
Firstpage :
201
Lastpage :
205
Abstract :
This paper presents a comprehensive load balancing approach using priority based migration (PBM) mechanism for a trusted small scale distributed computing environment. A process migration server (PMS) that acts as a cluster management server reduces latency time in migrated process execution; leading to no starvation condition for any process thus achieving uniform distribution of load among all the cluster nodes. The process to be transferred is selected on the basis of our proposed PBM approach and migration will be carried out between two nodes without intervention of PMS. Jingle-Mingle model reduces the network traffic by periodically broadcasting the node status and increases throughput by ensuring that maximum probability of process execution will be on the node, on which it has originated. To implement this mechanism, we use the hybrid and reliable load balancing algorithm with MOSIX as a middleware.
Keywords :
Internet; file servers; middleware; resource allocation; telecommunication traffic; Jingle-Mingle model; Mosix; broadcasting; cluster management server; cluster nodes; load balancing; middleware; network traffic; priority based migration; process migration server; trusted small scale distributed computing environment; Broadcasting; Delay; Distributed computing; Load management; Middleware; Network servers; Robustness; Telecommunication traffic; Throughput; Traffic control; Load Balancing; Network traffic; Process Migration; Process Migration Server;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Advances in Computing, Control, & Telecommunication Technologies, 2009. ACT '09. International Conference on
Conference_Location :
Trivandrum, Kerala
Print_ISBN :
978-1-4244-5321-4
Electronic_ISBN :
978-0-7695-3915-7
Type :
conf
DOI :
10.1109/ACT.2009.58
Filename :
5376750
Link To Document :
بازگشت