caGrid Overview

caGrid (or the Grid) is the underlying network architecture and platform that provides the basis for connectivity of caBIG^® tools. Connecting tools and sharing data across the Grid is the ultimate goal of the compatibility process. caGrid allows research groups to tap into the rich collection of emerging cancer research data while supporting their individual investigations. Key benefits include:

• Sharing of resources (including computational, storage, and data)
• Secure Access (including global authentication, local authorization, policies, and trust)
• Open Standards Virtualization

Quick Links to Related Resources

• Current Version of caGrid
• caGrid Knowledge Center
• caGrid Portal
• An Introduction to caGrid Technologies and Data Sharing: 16-page high-level overview of caGrid and its role in enabling data sharing. Developed for professionals who may oversee, plan or control data sharing within an institution, the publication covers topics including the purposes and uses of caGrid technologies, the use of structured descriptions of data to facilitate its exchange and use, and methods to control access to shared data.
• caCORE Overview.: Cancer Common Ontologic Representation Environment model drive architecture tools, caDSR (Cancer Data Standards Registry and Repository), and EVS (Enterprise Vocabulary Services), providing the critical elements needed to advertise and discover other Grid resources.

Getting on the Grid

To “get your tool on the Grid” you must: (1) Adopt or adapt a tool that is caBIG^® compatible; (2) Install caGrid software.

“Getting something on the Grid” – or “establishing a Grid node” – means meeting requirements for interoperability (through adoption or adaptation), and then connecting the interoperable service or data source to a Grid (internal or external). Grid services are “advertised” through an index. You can also connect to a Grid to “discover” service or data provided by others. All of the activities required to Grid-enable a system are faciliatated through caGrid tools and infrastructure. Adopting a caBIG^® compatible tool means much of the work has already been done; adaptation may take more effort. The activities required to Grid-enable a system (expose a compatible tool on the Grid) are:

• Use object types and information models registered in the Cancer Data Standards Registry and Repository (caDSR).
• Develop object oriented Application Programming Interfaces (APIs) and data resources.
• Define a Grid service interface to define the functionality that you are exposing to the Grid. This grid service interface uses the same object types as your existing system, but represents them in a way that is platform and language neutral (for example, using XML).
• Complete Grid service implementation by mapping service invocations to API calls or queries into the existing system.

Features of caGrid

caGrid has the following features:

• Nationally-deployed, standards compliant, data and analysis grid that any caBIG™-compatible system can plug into or draw from
• Semantically-discoverable and described services
• Rich runtime metadata infrastructure
• Uniform data query and retrieval across systems
• Federated security and identity management to support controlled access to systems
• Business Process Execution Language (BPEL) workflow engine for data analysis pipeline construction and execution
• Federated, cross-domain data mining and integration
• Standardized programming interfaces for application developers
• Graphical service development toolkit

The following diagram illustrates the components of caGrid: Community-Provided Services, Web Applications, Metadata Services, Security Services, Client Applications, and Business Activity Services.