CSE 262 Schedule

1/7/13 (Mon) Introduction

1/9/13 (Weds)
Technology

The future of microprocessors, by Shekhar Borkar and Andrew A. Chien, Comm. ACM 54(5):67-77 (May 2011).
Roofline: an insightful visual performance model for multicore architectures, S. Williams, A. Waterman and D. Patterson. Comm. ACM 52(4):65-76 (April 2009).
For background reading on memory hierarchies, including virtual memory, see one of
- Patterson & Hennessy, Computer Organization and Design: The Hardware/Software Interface, Patterson & Hennessy, Morgan Kaufmann, 3rd Edition
- What Every Programmer Should Know About Memory (Ulrich Drepper)

1/14/13 (Mon)
Algorithm based fault tolerance - linear algebra

Algorithm-based fault tolerance applied to high performance computing, G. Bosilca, et al., J Parallel and Distrib. Comput., 69(4):410-416 (April 2009).
Fault Tolerance in Message Passing Interface Programs, William Gropp and Ewing Lusk, Int. J. High Perform. Comput. Appl. 18(3):363-372 (August 2004).
J. Hursey and R. L. Graham, Building a Fault Tolerant MPI Application: A Ring Communication Example, 2011 IEEE Intl. Parallel Distributed Processing Symp., pp.1549-1556, 16-20 May 2011.

1/16/13 (Wed)
Resilience - iterative stencil methods

Fault resilience of the algebraic multi-grid solver, M. Casas, B. R. de Supinski, G. Bronevetsky and M. Schulz, Proc. 26th ACM Intl Conf Supercomputing (ICS '12), (2012).
For backround on Multigrid read Solving the Discrete Poisson Equation using Multigrid, by Jim Demmel, UC Berkeley. For background on solving Poisson's equation, read Solving the Discrete Poisson Equation, by Jim Demmel, UC Berkeley (Read through Jacobi's method).
For reference: Home page of the book A Multigrid Tutorial, Second Edition.

1/23/13 (Wed)
Architecture

1/28/13 (Mon)
Communication Avoiding Algorithms

Edgar Solomonik and James Demmel, Communication-optimal parallel 2.5D matrix multiplication and LU factorization algorithms, Tech Report UCB/EECS-2011-10, February 2011. Focus on matrix multplication; LU factorization could be discussed in another lecture, and is a good candidate reading.
Communication-Avoiding Parallel Strassen: Implementation and Performance , G. Ballard, J. Demmel, B. Lipshitz and O. Schwartz, Proc SC '12 (2012)

1/30/13 (Weds)
Project proposals

2/4/13 (Mon)
Communication tolerance

T. Nguyen et al., P. Cicotti, E. Bylaska, D. Quinlan and S. B. Baden. "Bamboo-Translating MPI applications to a latency-tolerant, data-driven form," SC12, 2012.

2/6/13 (Wed)
Energy Awareness

Dark silicon and the end of multicore scaling, H. Esmaeilzadeh et al., Proc. 38th Intl Symp Computer architecture (ISCA '11), 365-376 (2011).
See candidate papers list

2/11/13 (Mon)
Large graph problems

2/13/13 (Wed)
Sparse matrices

Autotuning Sparse Matrix-Vector Multiplication for Multicore, Jong-Ho Byun, Richard Lin, Katherine A. Yelick and James Demmel, UCB/EECS-2012-215, November 2012).
This paper builds on a body of prior work in the group on autuning.
- Eun-Jin Im, Katherine Yelick, and Richard Vuduc. Sparsity: Optimization framework for sparse matrix kernels. International Journal of High Performance Computing Applications, 18(1):135-158, 2004.
- R. Vuduc, J. W. Demmel, and K. A. Yelick. OSKI: A library of automatically tuned sparse matrix kernels. In Proc. SciDAC, J. Physics: Conf. Ser., volume 16, pages 521{530, 2005
- S. Williams, L. Oliker, R. Vuduc, J. Shalf, K. Yelick, and J. Demmel. Optimization of sparse matrix-vector multiplication on emerging multicore platforms. Parallel Computing, 35(3):178-194, 2009.
For more background, see candidate papers list

2/18/13 (Mon)

No class. President's day.

2/20/13 (Wed)
Project status reports

2/25/13 (Mon)
Parallel programming models

ispc: A SPMD compiler for high-performance CPU programming, by M. Pharr and W. R. Mark, Innovative Parallel Computing (InPar),, 2012.
To learn more about the practical details of ISPC, see Intel's github site
ISPC is also supported on Xeon Phi

2/27/13 (Wed)
Resilience

Fault tolerant preconditioned conjugate gradient for sparse linear system solution, M. Shantharam, S. Srinivasmurthy and P. Raghavan, 26th ACM Intl Conf Supercomputing (ICS '12), pp 69-78 (2012).

3/4/13 (Mon)
Language support

3/6/13 (Wed)
Large Graph Algorithms; automated performacne management

Erik Saule and U¨mit V. Catalyurek. An Early Evaluation of the Scalability of Graph Algorithms on the Intel MIC Architecture Proc. International Parallel and Distributed Processing Symposium Workshops and PhD Forum (IPDPSW '12), pp. 1629-39.
SEEC: A Framework for Self-aware Computing, H. Hoffmann, et al., Tech Report MIT-CSAIL-TR-2010-049, Mass. Inst Technology, 2010.

3/13/13 (Wed)
Final Project Presentations

3/15/13 (Fri) 12.00 to 2.00 PM - Room 2231
Final Project Presentations