Communications of the ACM
Daytona Project Will Help Researchers Crunch Data
Project Daytona will run a broad spectrum of machine-learning algorithms on the Windows Azure application hosting system.
Photo courtesy of Cloudtimes.org
Microsoft's Project Daytona, which will run a broad spectrum of machine-learning algorithms on the Windows Azure application hosting system, will benefit researchers seeking large-scale data computation and analysis capabilities.
The project is a component of the eXtreme Computing Group's Cloud Research Engagement Initiative, which debuted this week at the 12th annual Microsoft Research Faculty Summit.
Microsoft scientists pointed to high demand for a data-analysis and -processing architecture for education, health care, and environmental science workers who need tools to sort through large data repositories. "Daytona gives scientists more ways to use the cloud without being tied to one computer or needing detailed knowledge of cloud programming--ultimately letting scientists be scientists," says Microsoft's Dan Reed.
Powering Daytona is a runtime version of MapReduce, an open-license model produced by Google that can reach hundreds of server cores to analyze data. The MapReduce Runtime can be downloaded for free with sample codes and instructional materials to help researchers establish a cloud-analysis service on Azure, and user support and periodic updates also will be available.
From Sci-Tech Today
View Full Article
Abstracts Copyright © 2011 Information Inc. 
, Bethesda, Maryland, USA
No entries found