# It's Elemental

*Parallelization, targeting distributed memory architectures, of dense matrix computations is covered at least briefly in most introductory books and courses that include topics on numerical algorithms. The problem is that the algorithms that are typically covered are not those used in practice. The main objectives of this tutorial are to correct the basic misconceptions that have been perpetuated for at least two decades and to show how looking at the subject in just the right way exposes a systematic framework that allows novices to understand how we as experts develop and implement practical high performance libraries. This then allows us to bring participants to the forefront of the field, where new mechanical approaches automatically perform the tasks of the expert library developer in this domain.*

### Tutorial Information

International Conference on Supercomputing 2013Location: TBA

Times: 8:30-12:00 on Monday, June 10

Instructors:

- Bryan Marker (bamarker AT cs DOT utexas DOT edu)
- Jack Poulson (poulson AT stanford DOT edu)
- Robert van de Geijn (rvdg AT cs DOT utexas DOT edu)

- Main website is code.google.com/p/elemental
- Documentation available at poulson.github.com/Elemental
- Can check out with: hg clone http://code.google.com/p/elemental

#### A large collection of examples is available on GitHub

### Schedule

- Intro: 5 minutes
- Collective communication (van de Geijn): 15 minutes
- Parallel dense linear algebra (van de Geijn): 50 minutes
- Break
- Using Elemental (Poulson): 1 hour
- Design by Transformation (Marker): 1 hour

### References

#### Software

#### Poulson

- Communication lower bounds for distributed-memory matrix multiplication
- Communication-optimal parallel 2.5D matrix multiplication and LU factorization algorithms
- Elemental: A New Framework for Distributed Memory Dense Matrix Computations
- The fan-both family of column-based distributed Cholesky factorization
- I/O complexity: the red-blue pebble game
- Minimizing communication in numerical linear algebra

#### Marker

#### van de Geijn

- Collective communication: theory, practice, and experience
- Scalable Universal Matrix Multiplication Algorithms: 2D and 3D Variations on a Theme

© 2013 Jack Poulson

Template design by Andreas Viklund

Modified from Austin Benson's Introduction to Scientific Python