A Distributed Monitoring and Reconfiguration Approach for Adaptive Network Computing

The past decade has witnessed immense developments in the field of network computing thanks to the rise of the cloud computing paradigm, which enables shared access to a wealth of computing and storage resources without needing to own them. While cloud computing facilitates on-demand deployment, mob...

Full description

Saved in:
Bibliographic Details
Published in2015 IEEE 34th Symposium on Reliable Distributed Systems Workshop (SRDSW) pp. 31 - 35
Main Authors Bhargava, Bharat, Angin, Pelin, Ranchal, Rohit, Lingayat, Sunil
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2015
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:The past decade has witnessed immense developments in the field of network computing thanks to the rise of the cloud computing paradigm, which enables shared access to a wealth of computing and storage resources without needing to own them. While cloud computing facilitates on-demand deployment, mobility and collaboration of services, mechanisms for enforcing security and performance constraints when accessing cloud services are still at an immature state. The highly dynamic nature of networks and clouds makes it difficult to guarantee any service level agreements. On the other hand, providing quality of service guarantees to users of mobile and cloud services that involve collaboration of multiple services is contingent on the existence of mechanisms that give accurate performance estimates and security features for each service involved in the composition. In this paper, we propose a distributed service monitoring and dynamic service composition model for network computing, which provides increased resiliency by adapting service configurations and service compositions to various types of changes in context. We also present a greedy dynamic service composition algorithm to reconfigure service orchestrations to meet user-specified performance and security requirements. Experiments with the proposed algorithm and the ease-of-deployment of the proposed model on standard cloud platforms show that it is a promising approach for agile and resilient network computing.
DOI:10.1109/SRDSW.2015.16