grid computing

Sun's work on using grid computing to support high education

http://www.sun.com/products-n-solutions/edu/whitepapers/pdf/gridcomputing_architecture.pdf

CNGrid is a big Chinese project
WestGrid is a Canada Western site for grid computing

A data grid provides resources to locate, access, transfer, and consume data. Data grids can be used for many purposes, including:

• Building collaborative applications through data sharing. For example, a group of scientists studying the atmospheric ozone layer may collect huge amounts of experimental data per day. These scientists need efficient data storage and access to geographically dispersed storage facilities. Collaborative applications would support data sharing at various locations.
• Processing data using virtual resources such as CPU. In compute-intensive applications, such as the protein folding experiment, algorithm fragments could run independently on distributed resources capable of providing compute power. The data required for executing the algorithm and the resultant data need to be transferred from execution endpoint to other resources or monitoring applications.
• Providing on-demand computing on hosted database systems. SMB-scale applications can use on-demand data centers for hosting, accessing, and executing data through reliable, scalable, and secure mechanisms.

The core functional data requirements for these kinds of grid applications include:

• The ability to integrate multiple distributed, heterogeneous, and independently managed data sources
• Efficient data transfer mechanisms that provide data where the computation will take place, for better scalability and efficiency
• Data caching or replication mechanisms to minimize network traffic
• Data discovery mechanisms to let users find data based on data characteristics
• Data encryption and integrity checks to insure that data is transported across the network in a secure fashion
• Backup and restore mechanisms and policies necessary to prevent data loss and minimize downtime across the grid
• Security for data transport
• Distributed queries and filtering capabilities
• Proper utilization of the powerful capabilities of DBMSs to perform analytical and computational tasks such as complex query execution.