PCS201 PARALLEL AND DISTRIBUTED COMPUTING
Introduction: Scope and issues of parallel and distributed computing.
Models Of Parallel Computing: Taxonomy of parallel structures, Control mechanism, Address-Space Organization, Interconnection connection networks: Static and Dynamic interconnection networks, evaluating static interconnection networks, embedding other networks (Linear Array, Mesh, Binary Tree) into a hypercube; Routing mechanisms for static interconnection networks: Store and Forward (SF) Routing; Cut - Theory (CT) Routing; Cost-Performance trade-off; Architectural Models for Parallel Algorithm design.
Basic Communication Operation: Simple message transfer between two processors; One-to-all broadcast; All-to-all broadcast; Reduction and prefix sums; One-to-all personalized communication; All-to-all personalized communication; circular shift.
Performance And Scalability Of Parallel Systems: Performance matrices for Parallel systems: Run time, Speed up, Efficiency and Cost; The effect of granularity and data mapping on performance; Scalability of parallel systems; Iso-efficiency metric of scalability;
Models Of Distributed Computing: Mini computer model; Workstation pool model; Client-server model; Pool of processors model; Hybrid model.
Networking And Internetworking: Network technologies and Protocols.
Inter process Communication And Remote Procedure Calling: Building blocks; Client-server communication; group communication; Case study: Inter processor communication in UNIX; Design issues in Remote procedure calling; Implementation; Case Studies: SUN and ANSA;
Parallel Computing Algorithms: Various sorting and searching algorithms, performance metrics for parallel algorithm implementations.