document.write('
ABSTRACT



    While a system of n parallel processors is less efficient than one n-times faster processor, the parallel system is often cheaper to build. Parallel computation is used for tasks which require very large amounts of computation, take a lot of time, and can be divided into n independent subtasks. In simple terms parallel computing is the simultaneous execution of the same task (split up and specially adapted) on multiple processors in order to obtain results faster. The real strength of parallel computer is seen when an x86 machine is taken and attached with the group of parallel computer significant speedup is achieved in this manner. Tangible resources can be effectively utilized too by making use of parallel computer.

    In this project, we aim to demonstrate how significant speed up is achieved by using parallel computer. We have built two clusters, C5 which consist of five P IV nodes and Wall-O-Tile which is a supercomputer again a group of 5 nodes designed for multimedia applications. Linux Operating System is chosen as its\' stable system and gives pretty good performance. Also we have used some of the open source libraries to complete the project such as MPI, SDL, SMPEG.

    The C5 simply demonstrate the strength of parallel computer by solving many perplexing problems. These problem usually on a single computer can take hours, months or years to execute. Without doubt parallelized code is an efficient and cheaper alternative. We ran a series of program which tested the speed of these nodes ranging from a simple searching algorithm to complex brute forcing DES. Also we ran a standard benchmark test called LINPACK to evaluate the speed of our C5.

    The Wall-O-Tile is a solution for building a cost effective huge display system such as one found in football stadium, advertisement, etc. It is a solution for solving various scientific problem where minute microscopic details need to be studied. It is purely software based which makes it low-cost and highly efficient. It is very portable on the Unix-like environment as its\' cross-platform libraries. Also, it is an open source project that we can contribute to the society.

    Summing it up, the project is very challenging as we had to built a system from the scratch. Our achievements are (1) Building the institutes first cluster, (2) Running standard benchmarking tools, measuring the cluster speed and fine tuning it, (3) Building wireless cluster and comparing speed against wired cluster, (4) Writing parallel programs (5) Program to display images on tiled display (6) Program to run MPEG movie on tiled display , and (7) Creating a parallelized desktop for tiled display.   
');