Scalable data management for cloud services

Project Information

Discipline
Computer Science (401) 
Orientation
Research 
Abstract

In the context of the emerging cloud infrastructures, one of the most critical challenges concerns data management. Our work focuses on building an efficient and scalable storage service for IaaS Clouds by leveraging BlobSeer, the large-scale distributed data-sharing platform developed by the KerData INRIA team.

Intellectual Merit

We aim to evaluate the performance of different storage back ends for Cloud data services and to explore the ways to take advantage of the BlobSeer's scalable architecture, high throughput under heavy concurrency and versioning support to improve the Cumulus Cloud data service.

Broader Impacts

The goal of this project is developing new features for Cloud data services and improving the performance of data transfers in the Cloud.

Project Contact

Project Lead
Alexandra Carpen-Amarie (acarpena) 
Project Manager
Alexandra Carpen-Amarie (acarpena) 

Resource Requirements

Hardware System
  • hotel (IBM iDataPlex at U Chicago)
 
Use of FutureGrid

We plan to evaluate the Cumulus Cloud storage service through experiments that involve several distributed file systems.

Scale of Use

We wish to use about 20 nodes for our experiments. We occasionally may need to experiment with up to 100 nodes.

Project Timeline

Submitted
12/10/2010 - 12:36 
»