This project focuses on research issues involved in designing a document storage system over resources offered by a collection of peer-aware networked devices (in office environments). Increasingly typical office devices such as a Multi-Function Device (MFD) embed sophisticated compute power but are severely underutilized. Additionally most of them are connected to corporate network that are becoming faster and support increasing bandwidths. Thus the natural question is: can this available compute power be used to offer next generation of networked services?
This project would address the storage of documents over a grid of office devices. A device in a typical office environment may undergo variable load conditions and usage patterns that directly affect a devices contribution to storage grid. Hence, the device grid may be limited to transactional storage only. Such a document storage service can form the basis for many valuable services in office including Enterprise Information Portals and Content Management Systems.
This project also explores the use of cloud bursting to optimize service level agreements of near real-time data-intensive applications. Intercloud computation of data-intensive applications is a challenge due to large data transfer requirements, low intercloud bandwidth, and best-effort traffic on the Internet. The goal is to come up with scheduling heuristics as part of an autonomic cloud bursting approach that adapt to changing workload characteristics, variation in bandwidth, and available resources to optimize ordered throughput.
|Funding Agency:||XEROX Innovation Group|
|Principal Investigator:||Prof. D. Janakiram, IIT Madras|
|Collaborators:||Dr. Naveen Sharma and Dr. Nathan Gnanasambandam, Xerox|
|Students funded:||Sriram Kailasam (PhD. student), IIT Madras|