Economics Research Center / Computing Resources / Workshop Spring 2005
Spring 2005 Research Computing Workshop
Location: Rosenwald 329, 5:00 pm
| 1 | 3/31/05 | Introduction to Unix as a Working Environment [slides] |
| Beginning with a brief tutorial on the machines available and how to access them, this is an overview of basic Unix principles: shells, processes, filesystems, etc. Unix development toolchains will be introduced use in later sessions. | ||
| 2 | 4/7/05 | Matlab performance problems, shortcomings, and work-arounds [slides] |
| This is a quick run through of common performance problems with matlab, how to diagnose and then fix them. MEX will be discussed as well as common performance errors. | ||
| 4/7/05 | No meeting | |
| 3 | 4/21/05 | Bob Fourer: Mathematical Programming with AMPL [slides] |
| Description to be added | ||
| 4 | 4/28/05 | Jason Sarich: Network Enabled Optimization Solver (NEOS) [slides] |
| NEOS gives the entire world access to a large array of high quality commercial and free optimization solvers. Problems can be submitted remotely to a system of servers at Argonne National Lab and across the world. | ||
| 5 | 5/5/05 | C, the lingua franca |
| C is a "small" and extremely portable language which draws it's origins from system programming. Consequently it can be used to write programs that will run relatively efficiently on a wide variety of machines. While there are no scientific computing facilities built-in, there is wide support for it from the scientific computing community. | ||
| 6 | 5/12/05 | Parallel Computing with Message Passing -- MPI |
| The primary technique for performing parallel computation on distributed memory computers, message passing can be used to achieve significant performance increases on UofC servers. The Message Passing Interface (MPI) will be introduced and numerous examples will be given. Although an introduction, a familiarity with C would be beneficial. | ||
| 7 | 5/19/05 | Performance Optimization |
| Learn techniques to decrease program runtime and memory requirements. Common performance metrics and analysis tools will be reviewed as well as an emphasis on when optimization is and is not appropriate. Focus will be on imperative programs using a Unix-type toolchain. |
