Tag autotuning

Autotuning @ iWAPT’12

Speaker and audience at iWAPT'12 in Kobe, Japan

Live at iWAPT’12. Photo courtesy of Prof. Kengo Nakajima.

Rich is in Kōbe (神戸市), Japan to take part in the International Workshop on Automatic Performance Tuning (iWAPT), which is the 7th in this series of Japanese-run meetings. He had the honor of serving as this year’s iWAPT PC chair. Many thanks to co-chair Prof. Takeshi Iwashita and the program committee members for selecting a nice selection of papers.

CScADS’10 autotuning.

CScADS'10 Attendees in Snowbird, UT

Just returned from the DOE-sponsored CScADS Workshop on Libraries and Autotuning for Petascale Applications, held this year in the lovely Snowbird, Utah. The agenda featured exciting developments in domain-specific optimization, software engineering, compilers, languages, and mechanized derivation techniques. Click here for the full agenda and slides.

CScADS'10 attendees engaged in vigorous debate

Rich @ Snowbird for CScADS'10

CScADS Autotuning ’11

Rich is in Tahoe this week at the DOE CScADS Workshop on Libraries and Autotuning for Extreme-Scale Systems. The slides are being posted as the meeting progresses; you can follow the agenda here.

UPDATE: Slides for my talk posted here: [PDF]

Dagstuhl seminar.

Attendees of the Dagstuhl Seminar 10191 on Program Composition and Optimization

Nerds from the software engineering and the autotuning/HPC communities attempt to understand each other’s language and find common research ground at the Dagstuhl Seminar on Program Composition and Optimization (10191), at which Rich was an attendee.


It’s official: Rich’s NSF CAREER proposal on “Autotuning foundations for exascale systems” has been awarded. Thanks to everyone who provided commitments of support to the effort. I guess this means it’s time to get to work?

PPoPP’10 paper on GPU SpMV

Jee’s and Amik’s paper on autotuning sparse matrix-vector multiply for GPUs will appear at PPoPP 2010 in Bangalore, India. Great work, gang!

  • J. W. Choi, A. Singh, R. Vuduc. “Model-driven autotuning of sparse matrix-vector multiply on GPUs.” In Proc. ACM SIGPLAN Symp. Principles and Practice of Parallel Programming (PPoPP), Bangalore, India, January 2010. (accepted)

Autotuning @ Tahoe

Attendees of the CScADS'09 Autotuning Workshop in Tahoe

The US DOE-sponsored workshop on Libraries and Autotuning for Petascale Systems, held from Aug. 10–12 in Tahoe City, California, just completed. Jack Dongarra, Keith Cooper, Kathy Yelick, and Rich co-organize this annual research meeting. Perhaps not surprisingly, the most contentious discussion took place on the definition of autotuning!

The meeting slides are posted here: http://www.cs.utk.edu/~dongarra/cscads-libtune09.htm