[Menu Bar] Resourses at ARSC Science at ARSC Newsroom Support About ARSC ARSC Home

Symposium on Distributed Computing with MATLAB®

March 20-22, 2006

Butrovich 109

Register Online

The Arctic Region Supercomputing Center (ARSC) will be hosting a workshop on distributed computing with MATLAB, March 20-22, 2006, on the University of Alaska Fairbanks campus.

The Distributed Computing Toolbox (DCT) for MATLAB and the MATLAB Distributed Computing Engine (MDCE) have been available from the MathWorks for a little over a year, and are installed on various systems at ARSC. The workshop will begin with an overview of MATLAB fundamentals, suitable for anyone interested in knowing more about MATLAB components and uses. The first day will also feature an overview of DCT/MDCE, and details on how MATLABis available via site license to UAF.

The second and third days will feature hands-on instruction in the use of the DCT. Participants will be able to use MATLAB with DCT on their own portable computers, or ARSC's clusters. Instruction will progress from an overview of fundamental concepts and use through advanced parallel programming using DCT.

The DCT instruction featured in the workshop is not yet widely available, and represents the leading edge of MATLAB utilization for distributed processing, parallel algorithms, and other high-performance and high-throughput situations. The DCT is an important part of the evolution of MATLAB, giving extra appeal to MATLAB for analysis of large data sets, solutions for complex systems, or completion of many independent analyses.

Key Features of MATLAB®

  • High-level language for technical computing
  • Development environment for managing code, files, and data
  • Interactive tools for iterative exploration, design, and problem solving
  • Mathematical functions for linear algebra, statistics, Fourier analysis, filtering, optimization, and numerical integration
  • 2-D and 3-D graphics functions for visualizing data
  • Tools for building custom graphical user interfaces
  • Functions for integrating MATLAB® based algorithms with external applications and languages, such as C, C++, Fortran, Java, COM and Microsoft Excel

Pre-requisite knowledge: some experience with MATLAB. Advanced knowledge of MATLAB programming is desirable for MPI features of DCT (covered in Day Three). Note that ARSC is planning beginning and advanced MATLAB training for March 2-4.

Facilities: Desktop workstations with MATLAB will be available for participants to use during the workshop. Those with portable systems suitable for running MATLAB are encouraged to bring their own systems. Copies of MATLAB, including the DCT, will be available for use during the workshop.

Pre-registration is required, and space is limited.

To register please use our online registration form: Register Online.

Please email Dr. Greg Newby to for more information: . There is no fee for this workshop. Breaks and lunches are provided.

Schedule

Day One: Monday, March 20

9:00 - 9:15 Welcoming remarks and logistics (Frank Williams & Greg Newby, ARSC)
9:15 - 11:00 Overview of MathWorks, products and applications (Loren Dean, The MathWorks)
11:00 - 11:15
Break
11:15 - 11:30 "The Origins of MATLAB" video presentation by Cleve Moler
11:30 - 12:00 ARSC Data Center Tour (Greg Newby)
12:00 - 1:00
Lunch
1:00 - 2:00 MATLAB at UAF: getting the software, getting the licence, selecting the products, options (TBD) (DCT)
2:00 - 2:30 Overview of the Distributed Computing Toolbox (DCT) and MATLAB Distributed Computing Engine (Loren Dean, The MathWorks)
2:30 - 3:30 Open Questions & Answers about MATLAB, Simulink and related products; interactive demo of basic capabilities (John Kotwicki & Loren Dean, The MathWorks)
3:30
End of Day One
     

Day Two: Tuesday, March 21

8:30 - 9:00 Username logistics and account setup for ARSC systems (Derek Bastille, ARSC)
9:00 - 9:30 Day Two overview and logistics; system overview for hands-on users (Bastille, Newby)
9:30 - 10:30 Basic DCT/MDCE training
10:30 - 10:45
Break
10:30 - 12:00 Install/setup of DCT; Troubleshooting
12:00 - 1:00
Lunch
1:00 - 2:30 DCT Programming Considerations; Callbacks and Dynamic creation of tasks
2:30 - 2:45
Break
2:45 - 4:00 Scheduler concepts; pluggable schedulers; Q & A
4:00 End of Day Two
     

Day Three: Wednesday, March 22

9:00 - 10:00 Remarks by Cleve Moler, Chief Scientist of The MathWorks via remote video conference. Followed by Q & A
10:00 - 10:15 Recap and Q&A from Day Two; progress assessment, planning for the rest of the day
10:15 - 10:30
Break
10:30 - 11:00 ARSC systems for MATLAB (Tom Logan & Greg Newby, ARSC)
11:00 - 12:00 Using MPI: Parallel jobs with DCT/MDCE
12:00 - 1:00
Lunch
1:00 - 1:30 MPI in MATLAB, part II
1:30 - 2:30 Forthcoming DCT features, beta and beyond; distributed arrays, other schedulers, etc.
2:30 - 2:45 Moving forward: MATLAB in classes, MATLAB for research; attendee assessment of DCT, feedback on campus needs
2:45 - 3:00
Break
3:00 - 4:00 Open Q & A: Topics TBD
4:00
End of Day Three

 

For information on Fairbanks area hotels, contact

Other Training Events | ARSC Home

Arctic Region Supercomputing Center
PO Box 756020, Fairbanks, AK 99775 | voice: 907-450-8600 | email:

home | search | about | support | news | science | resources