BIRLA INSTITUTE OF TECHNOLOGY & SCIENCE, PILANI INSTRUCTION DIVISION FIRST SEMESTER 2005 - 2006 COURSE HANDOUT (PART II)

unevenoliveSoftware and s/w Development

Dec 1, 2013 (3 years and 6 months ago)

92 views

BIRLA INSTITUTE OF TECHNOLOGY & SCIENCE, PILANI

INSTRUCTION DIVISION

FIRST SEMESTER 2005
-

2006

COURSE HANDOUT (PART II)










Date: 02 / 08 / 2005

In addition to Part I (General Handout for all courses appended to the timetable) this portion gives fur
ther
specific details regarding the course.


Course No



:
CS/IS C422

Course Title


:
Parallel Computing

Instructor
-
in
-
charge

: T.S.B.SUDARSHAN

Email

:
tsbs@bits
-
pilani.ac.in
,

Course web page

:

http://csis/faculty/tsbs

Mirror Site

:

http://discovery.bits
-
pilani.ac.in/discipline/csis/tsbs/index.html


1. Course Description:


Introduction
to parallel computing; Models of parallel computers; Interconnection networks, basic
communication operations; Introduction to parallel algorithms; Parallel programming paradigms; issues in
implementing algorithms on parallel computers; Parallel programmin
g with message passing interface;
Performance analysis; Scalability analysis; Basic design technique for parallel algorithms; Parallel
algorithms for selected topics like sorting, searching and merging, matrix algebra, graphs.


2. Scope and Objective:




Parallel computing architectures have emerged as alternative to high performance computing
using powerful single processor machines. Sequential algorithms i.e., algorithms designed for single
processor machine, do not harness the full potential of a parall
el machine and hence the need to device
parallel algorithms. Parallel algorithms are highly architecture dependent. Moreover, for a given problem,
some parallel architectures are better suited than others. Therefore, it is necessary to study parallel
archi
tectures and techniques for designing efficient parallel algorithms.



3. Text Book:

Introduction to Parallel Computing

Ananth Grama, Anshul Gupta, George Karypis & Vipin Kumar, ,
Second Edition, Pearson Education, First Indian Reprint 2004.


4. Reference
Books:

(R1)

M.J. Quinn, “Parallel Computing : Theory & Practice’’, McGraw Hill Inc. 2
nd

Edition 1994.

(R2)

M.J.Quinn, “Parallel Programming in C with MPI and OPENMP”, Jaico Books, 2003

(R3)

F. Thomson Leighton, “Introduction to Parallel Algorithms and
Architectures : Arrays, Trees,
Hypercubes”, Morgan Kaufmann Publishers , 1992.

(R4)

IEEE Transaction on Computers

(R5)

IEEE Transaction on Parallel & Distributed systems.

(R6)

Journal of Parallel & Distributed Computing.


4. Course Plan:


Lecture
No.

Learn
ing Objectives

Topics to be covered

Reference to
text/ref. book

1





Introducing Parallel
Algorithms, studying
algorithms, minimizing number
Introduction to parallel processing and
parallel processing terminology

R1
Section
1.1,1.2 & 1.3

2,3

Contrast between Data Parallelism &
Control Parallelism

R1
Section 1.3 &
1.4

4
-
6

Parallel Programming Platforms

T1

Section 2.1


2.3.


7

of processors.

Physical Organization of parallel
computer

T1

Section 2.4

8,9

PRAM algorithms, (parallel reduction,
prefi
x sums, list ranking etc.)

R1

Section 2.3

10

PRAM Sorting Algorithms

Class Notes

11
-
13

Routing in parallel computer

T1

Section 2.5
-
2.6

14
-
16


Studying different
organizations, mappings
between them, data
decomposition and

Processor
-
Processor mapping
&
mapping techniques

T1

Section 2.7

17
-

19

Decomposition Techniques

T1

Section
3.1,3.2

20

Task Mapping

T1

Section 3.3

20,21



Performance metrics for
parallel systems

Performance metrics for parallel
system

T1

Section 5.1,
5.2

22,23

Isoefficeincy
function & scalability
issues

T1

Section 5.3,
5.4

24,25

Other scalability metrics

T1

Section 5.5,
5.6

26


Simple parallel algorithms on mesh
and hypercube

R1
Chapter 6

27
-
30



Studying & Analyzing parallel
versions of standard sequential
algorithms on
different
processor organizations

Parallel Matrix Algorithms: Matrix
Multiplication, Solving System of
Linear Equations

T1

Chapter 8

31
-
35

Sorting Algorithms on mesh and
hypercube

Class Notes +

T1

Chapter 9

36
-

40

Parallel Graph Algorithms: Prim’s,
Di
jkstra’s Algorithm etc.

T1
Section 10.1


10.4

The material in the text will be supplemented with a limited number of papers from IEEE transactions,
IEEE Computers and articles from the Internet, which describes the new machines. Some of these will be
mad
e available on the course web page.


5. Home assignments
will be announced in the class regularly.


6. Evaluation Scheme:


EC
No.

Evaluation
Component

Duration

(min)

Weightage

(%)

Date, Time & Venue

Nature of
Component

1

Test I

50

20

07/09 W 3
-
3:50 PM

220
7

Closed Book

2

Test II

50

20

16/11 W 3
-
3:50 PM

2207

Closed Book

3

Labs, Online and
Project*

-----

30

To be announced in
the class

Open Book

4

Comprehensive

180

30

7/12 (AN)

Partly Open


* Details of the assignments and project will be announced later.


8. Chamber Consultation Hours:

To be Announced

9. Make
-
up Policy:

Make Up for any component will be given only in genuine cases. In all cases prior intimation must be
given to IC.

10. Notices:

Notices regarding the course will be displayed on the notice
board of FD1.


















Instructor
-

in
-

charge











CS/IS C422