+ All Categories
Home > Documents > CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of...

CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of...

Date post: 03-Jan-2016
Category:
Upload: kerry-oliver
View: 221 times
Download: 3 times
Share this document with a friend
Popular Tags:
44
CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation
Transcript
Page 1: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

CS4402 – Parallel Computing

Lecture 1:Classification of Parallel Computers

Classification of Parallel Computation

Important Laws of Parallel Compuation

Page 2: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

How I used to make breakfast……….

Page 3: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

How to set family to work...

Page 4: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

How finally got to the office in time….

Page 5: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

What is Parallel Computing?

In the simplest sense, parallel computing is the simultaneous use of multiple computing resources to solve a problem.

Parallel computing is the solution for "Grand Challenge Problems“: weather and climate biological, human genome chemical and nuclear reactions

Parallel Computing is a necessity for some commercial applications: parallel databases, data mining computer-aided diagnosis in medicine

Ultimately, parallel computing is an attempt to minimize time.

Page 6: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 7: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 8: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

Grand Challenges Problems

Page 9: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

List of Supercomputers

Find this information at

http://www.top500.org/

Page 10: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

Reason 1: Speedup

Page 11: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

Reason 2: Economy

Resources already available. Taking advantage of non-local resources Cost savings - using multiple "cheap" computing resources instead of

paying for time on a supercomputer.

A parallel system is cheaper than a better processor. Transmission speeds. Limits to miniaturization. Economic limitations.

Page 12: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

Reason 3: Scalability

Page 13: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

13

Page 14: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

14

Types of || Computers

Parallel Computers

Hardware Software

Shared memory

Distributed memory

Hybrid memory

SIMD MIMD

Page 15: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 16: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 17: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

17

Page 18: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 19: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 20: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 21: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 22: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.
Page 23: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

The Banking Analogy

Tellers: Parallel Processors

Customers: tasks Transactions: operations Accounts: data

Page 24: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

Vector/Array

Each teller/processor gets a very fine-grained task

Use pipeline parallelism

Good for handling batches when operations can be broken down into fine-grained stages

Page 25: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

SIMD (Single-Instruction-Multiple-Data)

All processors do the same things or idle

Phase 1: data partitioning and distributed

Phase 2: data-parallel processing

Efficient for big, regular data-sets

Page 26: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

Systolic Array

Combination of SIMD and Pipeline parallelism

2-d array of processors with memory at the boundary

Tighter coordination between processors

Achieve very high speeds by circulating data among processors before returning to memory

Page 27: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

MIMD(Multi-Instruction-Multiple-Data)

Each processor (teller) operates independently

Need synchronization mechanism by message passing or mutual exclusion (locks)

Best suited for large-grained problems

Less than data-flow parallelism

Page 28: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

28

Important Laws of || Computing.

Page 29: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

29

Page 30: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

30

Page 31: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

31

Page 32: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

32

Page 33: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

33

Page 34: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

34

Page 35: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

35

Page 36: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

36

Important Consequences

f=0 when no serial part S(n)=n perfect speedup.

f=1 when everything is serial S(n)=1 no parallel code.

fn

nnS

)1(1)(

Page 37: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

37

Important Consequences

S(n) is increasing when n is increasing

S(n) is decreasing when f is increasing.

fn

nnS

)1(1)(

Page 38: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

38

Important Consequences

no matter how many processors are being used the speedup cannot increase above

Examples: f = 5% S(n) < 20 f = 10% S(n) < 10 f = 20% S(n) < 5.

ffn

nnS

1

)1(1)(

Page 39: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

39

Page 40: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

40

Page 41: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

41

Page 42: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

42

Gustafson’s Law - More

Page 43: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

43

Gustafson’s Speed-up

pnsT

TpnTs

TimeParallel

TimeSequentialnS

)(

snnsnsnS )1()1()( When s+p=1

Important Consequences:

1) S(n) is increasing when n is increasing

2) S(n) is decreasing when n is increasing

3) There is no upper bound for the speedup.

Page 44: CS4402 – Parallel Computing Lecture 1: Classification of Parallel Computers Classification of Parallel Computation Important Laws of Parallel Compuation.

44

To read:

1. John L. Gustafson, Re-evaluating Amdahl's Law,

http://www.scl.ameslab.gov/Publications/Gus/AmdahlsLaw/Amdahls.html

2. Yuan Shi, Re-evaluating Amdahl's and Gustafson’s Laws,

http://www.cis.temple.edu/~shi/docs/amdahl/amdahl.html

3. Wilkinson’s book,

1. sections of the laws of parallel computing

2. sections about types of parallel machines and compuation


Recommended