Pakistan Research Repository Home

Title of Thesis

Mohammad Ashraf Iqbal
Institute/University/Department Details
University Of Engineering & Technology/Electrical Engineering
Electrical Engineering
Number of Pages
Keywords (Extracted from title, table of contents and abstract of thesis)
load balancing strategies, partitioning problems, assignment problems, computer systems

Recent advances in VLSI (Very Large Scale Integration) and in automated design tools shave have removed a number of fundamental constraints from computer architecture. Complex processing logic is no longer costly. It is not essential to follow rules of von Neuman’s architecture in which a single processing unit operates serially on data items that it serially fetches from and restores to memory. It has become technically feasible to design computer structures in which tens of thousands of processors co-operate to solve problems at high speed. Parallel processing can be used to solve problems which can not be solved within an acceptable amount of time with the help of any commercial computer or which must be run on an expensive supercomputer, to which most researchers can only have access infrequently, if at all. The problems that are typically solved on parallel machines include linear system of equations, partial differential equations, signal processing, etc, which arise in many important scientific and engineering applications.

The performance of multiple processing system degrades if there is excessive inter communication between processors.

Download Full Thesis
1817.61 KB
S. No. Chapter Title of the Chapters Page Size (KB)
1 0 Contents
87.47 KB
2 1 Introduction 1
67.38 KB
3 2 A Comparative Analysis Of Static & Dynamic Load Balancing Strategies 8
338.88 KB
  2.1 Introduction
  2.2 Formulation of Problem
  2.3 The Optimal Static Algorithm
  2.4 The Binary Dissection Method
  2.5 The Greedy Algorithm
  2.6 The Predictive Dynamic Load Balancing Method
  2.7 Comparison of Results
  2.7 Conclusions
4 3 Approximate Algorithms For Partitioning And Assignment Problems 41
386.88 KB
  3.1 Introduction
  3.2 An Algorithm for Partitioning Chains
  3.3 Partitioning Multiple Chains across a Host-Setellite System
  3.4 Partitioning Distributed Programs in Host-Satellite System
  3.5 Partitioning Chains in Shared Memory System
  3.6 Partitioning Trees in Host-Satellite System
  3.7 Conclusions
5 4 Efficient Algorithms For Partitioning Problems 82
334.86 KB
  4.1 Introduction
  4.2 An Algorithm for Partitioning Chains
  4.3 Partitioning Multiple Chains across a Host-Satellite System
  4.4 Partitioning Trees in Host-Satellite System
  4.5 Conclusions
6 5 Partitioning And Assignment Problems For Binary Trees 115
377.87 KB
  5.1 Introduction
  5.2 Definitions
  5.3 An Scheme to Partition a Binary Tree
  5.4 Decomposition & Assignment Strategies
7 6 Efficient Algorithms For Dilated Mappings Of Binary Trees 154
344.35 KB
  6.1 Introduction
  6.2 Definitions
  6.3 An Algorithm to Find a Dilated-Mapping
  6.4 Applications
  6.5 Conclusions
8 7 Conclusions 192
66.94 KB