Enabling Managed Data Collaborations for eScience and eBusiness

A number of scientific collaborations funded by the DOE need to process an ever increasing volume of data. In the past, the primary challenge for these data intensive experiments was to move large data sets quickly from one place to another. With the adoption of grid standards, this challenge has been addressed. Today's challenge is to manage data collaboration, replication, and dissemination, with an emphasis on delivering end-to-end, managed quality of service that integrates seamlessly into current grid infrastructure. This project will extend the Globus Toolkit (GT), the de facto standard for grid computing middleware, and integrate it with new interfaces for managing the storage and network resources for the large data volumes that are characteristic of many of today's most exciting and promising science experiments. Phase I will determine what extensions are necessary to the Globus Toolkit, in order to enable a number of data collaboration scenarios that require end-to-end quality-of-service management. In particular, interfaces will be defined, prototypes will be developed, and specific functionality will be added. Commercial Applications And Other Benefits as described by the Applicant: The enhancement of the Globus Toolkit should enable end-to-end quality-of-service management for data collaboration, allowing the efficient cooperation and integration of computing resources, network bandwidth, and data storage facilities. Put another way, scientists, engineers, and even everyday workstation users will be able to more transparently work with and manage huge volumes of data, allowing each to focus on the science, engineering, or business at hand rather than the mundane details of managing terabytes or even petabytes of data. Both government and private businesses should realize substantial productivity gains and the economic benefits of those gains.