CLUSTER TECHNOLOGIES Foilene ble ogs presentert p NOTUR 2003

1
CLUSTER TECHNOLOGIES(Foilene ble også
presentert på NOTUR 2003)

• Anne C. Elster
• Dept. of Computer Information Science (IDI)
• Norwegian Univ. of Science Tech. (NTNU)
• Trondheim, Norway

NFR 30. juni, 2003
2
Clusters (Networks of PCs/Workstation)

• Are they suitable for HPC?
• Advantage
• Cost-effective hardware since uses COTS
  (Commercial Of-The-shelf) parts
• BUT
• Typically much slower processor interconectes
  than traditional HPC systems
• What about usability?

NTNU IDIs 40-node AMD 1.46GHz cluster 2GB RAM,
40GB disk, Fast Ethernet
3
Cluster TechnologiesNOTUR Emerging Technology
projectCollaboration between NTNU Univ. of
Tromsø

• Goal
• Analyze Cluster technologies suitability for HPC
  by looking at some of the most interesting NOTUR
  applications
• The results will provide a foundation for
  decisions regarding future HPC programs

4
Main Collaborators include

• Anne C. Elster (IDI, NTNU) Project leader
• Otto Anshus Tore Larsen (CS, U of Tromsø
• Torbjørn Hallgren (IDI, NTNU)
• Einar Rønquist (IMF, NTNU)
• Master , Ph.D. Students Post Docs at NTNU and
  Univ. of Tromsø

5
General Issues to Consider

• Why cluster vs. Powerful desktop vs. Large SMPs?
• What are the total costs associated with clusters
  (harware, software, support, usability)
• 32-bit vs. 64-bit architectures

6
Cluster Project ACTIVITIES

• Profiling Tuning Selected Applications
• Physics and Chemistry Codes
• (Elster students, Dept. of Computer Science
  Dept., NTNU)
• Profiling User-Analysis of Amber, Dalton
  Gaussian
• (Tor Johansen staff, Comp. Center, U of
  Tromsø)
• Optimization tool analysis of Dalton
• (Anshus PostDoc/student, Dept. of Comp. Sci.,
  U of Tromsø)

7
Cluster Project ACTIVITIES continuted

• Execution Monotoring
• (Anshus, Tore Larsen students, CS, U of T)
• Visualization servers, etc.
• (Hallgren, Elster students, CS, NTNU)
• Impact of future numerical algorithms
• (Rønquist student, Dept. of Mathematics, NTNU
• Interface with NOTUR ET Grid Project
• (Elster, Harald Simonsen and colleagues, staff
  students associated with the NOTUR ET Cluster
  Grid projects)

8
Some cluster issues discovered

• Performance of programs can individually vary on
  different machines
• FORTRAN problems
• Different FORTRAN implementations have
  non-stardard add-ons (e.g. FORTRAN 90)
• Leads to great difficulty in porting code to a
  different platform with a different Fortran
  compiler (e.g. by a different vendor)

9
Some cluster issues discovered continued

• Global operations have more severe impact on
  performance on clusters than traditional
  supercomputers since communication between
  processors take relatively more time of total
  execution time

10
The Ideal Cluster -- Hardware

• High-bandwidth network
• Low-latency network
• Low Operating System overhead (tcp causes slow
  start)
• Great floating-point performance (64-bit?)

11
The Ideal Cluster -- Software

• Compiler that is
• Portable
• Optimizing
• Do extra work to save communication
• Self tuning /Load ballanced
• Automatically choose best algorithm
• One-sided communication support?
• Optimized middleware

12
For more information

• Poster session at NOTUR
• Cluster Project stand
• - several posters smoke simulation demo
• Poster by Torbjørn Vik et. al.
• Poster by Lars Ailo Bongo
• Email elster_at_computer.org