Overview of Supercomputers

1
Overview of Supercomputers

• Presented by
• Matt Lockwood
• Mohammad Haghkar

2
Supercomputers

• The category of computers that includes the
  fastest and most powerful (most expensive) ones
  available at any given time.
• Designed to solve complex mathematical equations
  and computational problems very quickly.

3
What are They Used For

• Climate prediction Weather forecasting

4
What are They Used For (cont.)

• Computational chemistry
• Crash analysis
• Cryptography
• Nuclear simulation
• Structural analysis

5
How Do They Differ From a Personal Computer

• Cost
• range from 100,000s to 1,000,000s
• Environment
• most require environmentally controlled rooms
• Peripherals
• lack sound cards, graphic boards, keyboards, etc.
• accessed via workstation or PC
• Programming language
• FORTRAN

6
History

• Seymour Cray (1925-1996)
• Developed CDC 1604 first fully transistorized
  supercomputer (1958)
• CDC 6600 (1965), 9 MFlops
• Founded Cray Research in 1972 (now Cray Inc.)
• CRAY-1 (1976), 8.8 million, 160 MFlops
• CRAY-2 (1985)
• CRAY-3 (1989)

7
Early Timeline of Supercomputers

8
Where Are They Now

• www.top500.org
• List released twice a year
• Scores based on Linpack benchmark
• Solve dense system of linear equations
• Speed measured in floating point operations per
  second (FLOPS)

9
Architectures - SMP

• Symmetric Shared-Memory Multiprocessing (SMP)
• Share memory
• Common OS
• Programs are divided into subtasks (threads)
  among all processors (multithreading)

10
Architectures MPP

• Massively Parallel Processing (MPP)
• Individual memory for each processor
• Individual OSs
• Messaging interface for communication
• 200 processors can work on same application

11
Architectures Clustering

• Grid computing
• Many servers connected together
• Relies heavily on network speed
• Easily upgraded with addition of more servers

12
Processor Types

• Vector processing
• Expensive
• NEC Earth Simulator
• Scalar processing
• Grid computing
• Based on off the shelf parts (ordinary CPUs)

13
BlueGene/L

• IBM
• MPP (massively parallel processing)
• 1 on top500 as of November 2004
• 32,768 processors (700Mhz)
• 70.72 Teraflops (trillions of FLOPS)
• Runs linux
• DNA, climate simulation, financial risk
• Cost more than 100 million

14
BlueGene/L System Layout

• 2 Processors
• Node communication
• Mathematical calculations

15
BlueGene/L Compute Card
16
BlueGene/L Node Board
17
BlueGene/L Cabinet
18
Some of the Others

• 2 - Columbia (NASA, USA) 51.87 TFlops
• 3 - Earth Simulator (Japan) 35.86 TFlops
• 4 - MareNostrum (Spain) 20.53 TFlops
• 5 - Thunder (USA) 19.94 TFlops

19
The Future
20
References

• http//www.top500.org/
• http//www.pcquest.com/content/Supercomputer/10205
  1004.asp
• http//news.com.com/2100-1008_3-1000421.html?tagf
  d_lede2_hed
• http//www.research.ibm.com/bluegene/index.html
• http//www.llnl.gov/asci/platforms/bluegene/papers
  /2hardware_overview.pdf
• http//www.hpce.nec.com/451M5f7cd421b8e.0.html
• http//www.cray.com/about_cray/history.html
• http//www.serc.iisc.ernet.in/govind/243/L7-PA-In
  tro.pdf
• http//www.computerworld.com/hardwaretopics/hardwa
  re/server/story/0,10801,43504,00.html