Modern data processing tasks involving high computation with huge data intensive work are not providing any usual response as they run over a conventional computing architecture, where the synergism capabilities of such machines are limited to single central processing unit. Improvement over such single processing architecture is not the big issue as many earlier efforts in this era has been performed, which involves overlapped pipelined architectures. Later the technology extends to involve multiple processing elements under the control of a common clock. A current trend involves multiple central processing units. Despite of such efforts, another way of achieving parallel effect is to make effective utilization of multi-computer hardware in the form of massively parallel clustering over a local area network. Further the experiment lead to the analysis of Run-length image compression over a network cluster-involving client – server model of computation consisting software modules implemented via TCP/IP sockets for the requirement of increased speedup as well as throughput. Finally, the conclusion containing comparisons over clustered environment will be discussed.

Parallel Clustering Multi-Computers Run-length Image Compression.