Network Computing

• The Internet is the most widely used distributed system in the world and has stimulated rapid growth of interest in largescale network computing that may span the entire globe. Nodes in the Internet may be
  • single-processor workstations,
  • shared-memory MIMD machines,
  • massively parallel SIMD machines, or other types.
• Links are TCP/IP packet-switched connections and the bandwidth varies with load, number of hops, and underlying communication technology. When one node connects with another, the packets of data may be sent through
  • a wireless link,
  • fiber optical cable,
  • coaxial cable,
  • digital telephone line, and so on.
• These physical layers introduce delays and may be errors, which must be corrected by retransmission and dynamic reconfiguration of the Internet's links.
• In network computing, the nodes are stand-alone computers that could be connected
  • via a switch,
  • local area network,
  • the Internet.
• The main idea is to divide the application into semi-independent parts according to the kind of processing needed. Different nodes on the network can be assigned different parts of the application.
• This form of network computing takes advantage of the unique capabilities of diverse system architectures. For example, the fine-grained SIMD part of the application would be shipped off to the SIMD machine and the graphical presentation and I/O portions to one or more single-processor workstations.
• It also maximally leverages potentially idle resources within a large organization. Therefore, unused CPU cycles may be utilized during short periods of time resulting in bursts of activity followed by periods of inactivity (the utilization of network technology in order to create a computing infrastructure using commodity computers).