Home Verilog Digital Design Digital Basics Python RF Basics

Legal Disclaimer

Chip Designing for ASIC/ FPGA Design engineers and Students
Digital-logic Design...  Dream for many students… start learning front-end…
Custom Search

Feedback ? Send it to admin@fullchipdesign.com or join me at fullchip@gmail.com

Digital Design Clock Crossing Async FIFO Half Adder Full Adder Binary Adder Overflow Overflow Det Adder-Subtractor Multiplier Parity check RTL guidelines NAND to INVERTER VHDL RTL Arith Micro-ops Stack Org Parallel proc. Pipeline proc CMOS Intro

Legal Disclaimer

Topics @TYH :- 4G LTE Tutorial, GVIM editor, Smart-Phone, Cloud Computing
Previous Next
Verilog Tutorial.
Digital Basics Tutorial.
Parallel Processing.  Next topic Pipeline Processing
Parallel Processing is an implementation where data is computed simultaneously using multiple instantiation of same digital circuit.

Faster execution time., so higher throughput.  

More hardware required, also more power requirements. Not good for low power and mobile devices.

There are different ways to classify parallel processing in modern computers.  Some of the most common approaches for classification are SISD, SIMD, MISD and MIMD.
According to M..J Flynn.  Most computers fetch instructions from memory and it constitutes  instruction stream (IS). The instructions are decoded and results into fetch of data. This second category is called data stream (DS) which involves operations on data.

Flynn's classified digital computers in following categories:-

1. SISD - Single instruction stream and single data stream.

In this architecture only 1 processor,  1 memory and 1 control units are implemented. In this 1 instruction can be executed at any time. So the instructions are held at input and serially pushed one after another.   

2. SIMD - Single instruction stream and multiple data stream.

In this architecture many processor’s are implemented to work with single control unit. The multiple memory module needs to be implemented to support each control unit independently.  

3. MISD - Multiple instruction stream and single data stream.

Theoretically this architecture implements many processor’s to work with independent control units. The single memory module is implemented to support only one data path. Not logical to implement practically as a product.    

4. MIMD - Multiple instruction stream, multiple data stream.

In this architecture, many processor’s are implemented to work with multiple control units to process multiple instructions in parallel. The multiple memory modules are implemented to support parallel data streams.  
According to Flynn, we can classify Pipeline processing as an efficient implementation of hardware to process multiple instructions
Interview Questions. Main, FPGA, Digital Fundamentals
LTE - Long Term Evolution topics from here