A physical view
- 1. Prof. Neeraj Bhargava Pooja Dixit Department of Computer Science School of Engineering & System Sciences MDS, University Ajmer, Rajasthan, India 1
- 2. A grid is a collection of networks, processors, storage, and other resources. Networks The networking mechanism is the most fundamental resource for the grid and also is the theme of this book. In fact, without networking grid computing would not be possible. The recent growth in communication capacity makes grid computing practical, compared to the limited bandwidth available when distributed computing was first emerging. Transmission of content and job supervision within the grid are important for sending jobs and the required data to points within the grid (some jobs require a large amount of data to be processed and it may not always reside on the processor running the job. 2
- 3. Computation The next most common resource on a grid is obviously computing cycles provided by the processors on the grid. The processors can vary in speed, architecture, software platform, and storage apparatus. There are efforts underway to develop very high- speed supercomputers. Whereas clustering is a common approach at the TFLOPS speeds, grid computing can also play a role in these initiatives by refining architectures that link remote computers into an assembly of loosely or tightly coupled processors. At the business level, grid computing is expected by the industry to be more practical than cluster-based supercomputing. 3
- 4. Storage The next most common resource used in a grid is data storage. In a grid environment, a file or data base can span several physical storage devices and processors, bypassing size restrictions often imposed by file systems that are preembedded with operating systems. Storage capacity available to an application can be increased by making use of the storage on multiple processors with a unifying file system. Each processor on the grid usually provides some quantity of storage for grid use. Storage can be “primary storage,” “secondary storage,” or “tertiary storage.” Memory directly attached to a processor has fast access capabilities but is volatile; this kind of memory is used to cache data to serve as temporary storage for running applications. “Secondary storage” is generally implemented in hard disk drives, such as with RAID (redundant array of inexpensive drives). “Tertiary storage” is generally implemented in near-real-time accessible media such as tape or other permanent storage media. Many grid systems use mountable networked file systems, such as Network File System (NFS), Distributed File System (DFS), or General Parallel File System (GPFS). Special grid database software can “federate” a group of individual data bases and files to form a larger, more inclusive data base. A PHYSICAL VIEW 4
- 5. Scientific Instruments Grids, particularly an intergrid, can provide shared access to expensive scientific equipment or interconnect dispersed equipment into a larger overall scientific tool. We will not discuss this issue further, since our emphasis is on commercial applications. Software and Licenses Two issues are of interest to organizations: application software and licensed software. Regarding application software, the most basic approach is to use the grid environment to allow the application to run on an available processor on the grid (rather than running locally.) Further along the transition trajectory, one can modify the application to segment its work in such a way that the separate parts can execute in parallel on different grid processors. “scalability” is a measure of how efficiently the multiple processors on a grid are used. However, there may be limits to scalability. A PHYSICAL VIEW 5