mca Syllabus

Advanced Computer Architecture
Code: PGCS202
Weekly Contact Hour: 3L
Credit: 3
Course Contents

Introduction to High Performance Computing: Overview, Pipeline vs Parallel Processing Parallel Architectures: Classification and Performance. Pipeline Processing: Pipeline Performance, design of arithmetic pipelines, multifiction pipes, concept of reservation table, collision vector and hazards.
Instruction Processing Pipes: Instruction and data hazard, hazard detection and resolution, delayed jumps, delayed execution. RISC Philosophy. Pipeline scheduling Theory: Greedy pipeline scheduling algorithm, state diagram, modified state diagram, Latency cycles, Optimal cycles, scheduling of static & dynamic Pipelines. Implementation of pipeline schedulers Interconnection Networks: Interconnection network classification, Single stage/ Multistage Networks, crossbars, clos Networks, Benes Networks, Routing algorithms. Omega, Cub-connected and other networks. Introduction to Neurocomputing Architectures Topics from the current literature as self study and presentations by students.

Books
1. M. R. Bhujade, "Parallel Computing", Newage International Pvt. Ltd., 1995.
2. Stallings, William, "Computer organization and architecture, designing for performance", Prentice Hall of India, 1997
3. J. L. Hennessy and D. A. Patterson, "Computer architecture: a quantitative approach", Harcourt Asia, Singapore 1996
4. Kain, “Advanced Computer Architecture: a system Design approach”, PHI.