Unit 2: Grid terms and concepts
Grid computing is a system where many computers are connected to share resources and work together as one powerful system. These computers may belong to one or different organizations, and some resources may have restrictions.
The main resource in a grid is computation (CPU power). Applications can run on remote machines, be divided into parallel parts, or run many times on different machines. Scalability means that adding more processors reduces execution time.
The second resource is storage. Storage can be memory (temporary) or hard disk (permanent). Grid file systems combine storage from many machines and allow users to access data easily. Techniques like data striping improve speed and performance.
Communication is also important because machines must transfer jobs and data. Network speed affects performance.
The grid also manages software and licenses, sending jobs only to machines with required software. Some machines may have special hardware or follow specific policies.
Work in a grid is divided into applications and jobs. The grid uses scheduling, reservation, and scavenging to assign jobs efficiently.
Grids can exist within one organization (intragrid) or between many organizations (intergrid).
Grid computing combines computing power, storage, communication, and software from many machines to complete large tasks efficiently.