Chapter 9, Problem 7RETC

Explanation of Solution

Instruction level parallelism:

It is one of the ways by which many computer operations could be performed at the same time which means multiple instructions can be executed simultaneously. Even if the instructions are written simultaneously and are to be executed sequentially, but many of these instructions are not dependent on each other, still they give correct solutions when executed simultaneously. From this we can derive that the compiler can execute them together, not sequentially which saves time and yields faster results. The different ways by which instruction level parallelism was exploited are the instruction pipelining, superscalar execution, register renaming, branch prediction, and out of order execution.

Need for innovation of instruction level parallelism:

Since the invention of computers, scientists have tried their best to make these computers better, faster and more efficient in solving the most critical problems, to deal with complex problems and large junks of data with minimal space. However they always face the physical and economical restrictions. Therefore, the solution derived to enhance the performance of the processor was not by increasing the number of processors but by finding some better intelligent technical solutions for the instructions codes and one of the ways was parallelism...