Recent Parallel Processing Computer System Technology

Note

This is one of my Doctoral assignment from Current Science and Technology in Japan Course which has never been published anywhere and I, as the author and copyright holder, license this assignment customized CC-BY-SA where anyone can share, copy, republish, and sell on condition to state my name as the author and notify that the original and open version available here.

Pipelining for microprocessor

A microprocessor is an electronic component that is used by a computer to do its work. It is a central processing unit (CPU) on a single integrated circuit (IC) chip containing millions of very small components including transistors, resistors, and diodes that work together [1]. The traditional microprocessor is too simple, but it is good to be explained in the class. The traditional one have five instructions which in order are fetch, decode, execute, memory, and write. In Figure 1 is seen that the program counter accesses the instruction memory, then the register fetch the instruction, next the instruction is decoded by the decoder, later it is sent to the arithmetic logic unit (ALU) and execute the instruction, finally the result is stored in the data memory and written into the register [2].

Figure 1. Simple Microprocessor Diagram

Pipelining is a technique to speed up the processing. Without pipelining the processor will have to wait until the whole 5 steps finishes before it can execute a new one, in other words serial processing. However pipelining allows the processor to start processing the next instruction without waiting until the previous instruction processing is finished, in other words parallel processing. Figure 2 showed the simplest illustration, but the technique have grown vast, for examples there are parallel operation, superscalar, super pipelining, and very long instruction word (VLIW). There are also data dependency problems such as flow, control, anti, intput, and output that prevents performance improvement. There are some…