1. 1 Parallel Computing@University of Hyderabad CS-726 Parallel Computing By Rajeev Wankar wankarcs@uohyd.ernet.in

2. 2 Parallel Computing@University of Hyderabad For whom  Elective for M.Tech. and MCA

3. 3 Parallel Computing@University of Hyderabad Objective By the end of the semester, students should be able to develop the following skills:  Should be able to understand parallel algorithm paradigms and design efficient parallel algorithms.  Given a practical application, identify design issues and should be able to write algorithms for the targeted machine.  Develop skill to write/modify parallel library.

4. 4 Parallel Computing@University of Hyderabad Prerequisite Knowledge of Introductory Algorithms, Networks, Java/C/C++, and Unix/Linux (as an OS and good if you know socket programming).

5. 5 Parallel Computing@University of Hyderabad Course Outline Here is a preliminary and non-exhaustive list of topics we will be or might be covering. This is subject to change with advanced notice, partly based on the understanding of the students.

6. 6 Parallel Computing@University of Hyderabad Unit 1  Introduction to Parallel Computing:  Why Parallel Computing & Scope of Parallel Computing, Control and Data Approach, Models of parallel computation, Design paradigms of Parallel Algorithms.

7. 7 Parallel Computing@University of Hyderabad Unit 2  Classification: Taxonomies:  MPP, SMP, CC-NUMA, cluster: dedicated high performance (HP), high throughput (HT), data-intensive computing, Interconnection networks, Flynn‘s Taxonomy.

8. 8 Parallel Computing@University of Hyderabad Unit 3  An overview of Practical Parallel Programming Paradigms:  Programmability Issues, Programming Models: Message passing, client-server, peer-to-peer, Map & Reduce.

9. 9 Parallel Computing@University of Hyderabad Unit 4  Clustering of Computers, Beowulf Supercomputer, Use of MPI in Cluster Computing. Debugging, Evaluating and tuning of Cluster Programs

10. 10 Parallel Computing@University of Hyderabad Unit 5  Message passing standards:  PVM (Parallel Virtual Machine), MPI (Message Passing Interface) Message Passing Interface (MPI) and its routines.

11. 11 Parallel Computing@University of Hyderabad Unit 6  Performance Metrics & Speed Up:  Types of Performance requirements, Basic Performance metrics; Workload Speed Metrics; Performance of Parallel Computers-Parallelism and interaction overheads;

12. 12 Parallel Computing@University of Hyderabad Unit 6  Overview of Programming with Shared Memory:  OpenMP (History, Overview, Programming Model, OpenMP Constructs, Performance Issues and examples, Explicit Parallelism: Advanced Features of OpenMP)  Distributed Shared Memory programming using Jackal  Introduction to “Programming Multi-Core Programming” through Software Multi-threading

13. 13 Parallel Computing@University of Hyderabad Unit 7  Reconfigurable Computing  What is it? Why? How to do it? Where to do it?  Algorithms for Reconfigurable systems

14. 14 Parallel Computing@University of Hyderabad Unit 8 (Applications)  Built cluster using Rocks  On Cluster Based algorithms and applications  On Shared Memory Programming  Writing subset of parallel libraries using socket programming in C or Java.

15. 15 Parallel Computing@University of Hyderabad Assessment  Internal: 40 Marks  Three internals of 10 marks each (best two of three will be selected)  Lab assignments 10 marks  One Group assignment 5 marks  Seminar 5 marks  External: End semester examination 60 Marks.

16. 16 Parallel Computing@University of Hyderabad Reference:  Quinn, M. J., Parallel Computing: Theory and Practice (McGraw-Hill Inc.).  Bary Wilkinson and Michael Allen: Parallel Programming Techniques using Networked of workstations and Parallel Computers, Prentice Hall, 1999.  R. Buyya (ed.) High Performance Cluster Computing: Programming and Applications, Prentice Hall, 1999.  William Gropp, Rusty Lusk, Tuning MPI Applications for Peak Performance, Pittsburgh (1996).  Kai Hwang, Zhiwei Xu, Scalable Parallel Computing (Technology Architecture Programming) McGraw Hill Newyork (2004).  W. Gropp, E. Lusk, N. Doss, A. Skjellum, A high performance portable implementation of the message passing Interface (MPI) standard, Parallel Computing 22 (6), Sep 1996.  Gibbons, A., W. Rytter, Efficient Parallel Algorithms (Cambridge Uni. Press).  Kumar V., et al., Introduction to Parallel Computing, Design and Analysis of Parallel Algorithms, Benjamin/Cummings, 1994.  Shameem A and Jason, Multicore Programming, Intel Press, 2006

17. 17 Parallel Computing@University of Hyderabad Questions