Grid Computing

• While clusters are collections of computers tied together as a single system,
• grids consist of multiple systems that work together while maintaining their distinct identities.
• Owing to the decentralized and heterogeneous nature of the grid,
• the middleware that glues the different components is more complicated compared with that of clusters.
• Resembling an electric power grid, the computing grid is expected to become a pervasive (spread throughout) computing infrastructure that supports large-scale and resource-intensive applications.
• The significant increase in application complexity and the need for collaboration have made grids an attractive computing infrastructure.

• Thus, the need for the distributed grid infrastructure will continue to be an important resource.
• A user signing on at one location would view computers at other remote locations as if they were part of the local system.
• Grid computing works by polling the resources available,
• and then allocating them to individual tasks as the need arise.
• Resources are returned to the pool upon completion of the task.
• Grid gives an illusion of a big virtual computer capable of carrying out enormous tasks.

• Support of grids requires innovative solutions to a number of challenging issues including:
  • resource management,
  • resource monitoring,
  • interoperability,
  • security,
  • billing and accounting,
  • communication, and
  • performance.

• There are several examples of grid platforms and tools such as Globus and TeraGrid.
  • The Globus Toolkit is an enabling technology for the grid.
  • The toolkit includes software services and libraries for resource monitoring, discovery, and management, plus security and file management.
  • It also includes software for communication, fault detection, and portability.
  • The Globus Toolkit has grown through an open-source strategy. Version 1.0 was introduced in 1998 followed by the 2.0 release in 2002. The latest 3.0 version is based on new open-standard Grid services.

• TeraGrid is a large high-performance computing project headed by the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign.
• The TeraGrid uses thousands of Intel Itanium 2 processors located at four sites in the United States.
• The TeraGrid is an effort to build and deploy the world's largest, fastest distributed infrastructure for open scientific research.
• The TeraGrid is expected to include 20 teraflops of computing power, facilities capable of managing and storing nearly 1 petabyte of data, high-resolution visualization environments, and toolkits for grid computing.
• These components will be tightly integrated and connected through a network that will operate at 40 gigabits per second.