Category Talks


The HPC Garage is in Savannah, GA this week for the SIAM Parallel Processing (PP) meeting. We are presenting and co-organizing in various minisymposia, so please drop by!

  • Minisymposia, co-organized by Rich: Energy-aware high-performance computing: Part 1 (MS2) and Part 2 (MS8)
  • Talk: Model-based hybrid scheduling for the fast multipole method, by Jee Choi. In MS18: Multicore, GPU, and GPU+CPU hybrid implementations of CFD methods.
  • Talk: Intra-node performance tuning and analysis of fast n-body methods from multicore to exascale. In MS30: Directed acyclic graph approaches for parallel scientific computing.
  • Talk: CnC for HPC? by Rich. In MS42: Parallel programming models, algorithms, and frameworks for scalable manycore systems.


Taking in the view at ICERM (Institute for Computational and Experimental Research in Mathematics), just before the start of a meeting on exascale algorithms, architectures, and programming [www]. I (Rich) gave a talk on Kent’s preliminary work that is developing a power- and area-constrained model for co-designing algorithms and architectures. Never before have I heard references to work done “30 years ago” than at this meeting, at which very talented, experienced, and knowledgeable computational scientists abound.


Cong is giving a talk on program inversion this week at the 16th Workshop on Compilers for Parallel Computing (CPC) [www].

UPDATE: Cong’s slides are here: [PDF]

Aparna’s talk at SC’11

Aparna gave a great talk on the future of the fast multipole method! I only regret that the SC’11 organizers messed up the title on the billboard outside the room. (It has the title of the talk as, “Performance tuning and analysis of the fast Aparna Chandramowlishwaran multipole method from multicore to exascale.” It remains to be seen if this “FACMM” algorithm will catch on, though it certainly deserves to.)

GT @ SC’11

The HPC Garage is at Supercomputing this week! A few notable lab-specific activities appear below, but please go to to see even more of how Georgia Tech represents.

  • Poster: Prospects for scalable 3D FFTs on heterogeneous exascale systems (Kent, Casey, Chris, Karthik, P.-K. Yeung, Rich), Tuesday Nov 15 5:15-7pm [www]
  • Talk: George Michael Memorial Fellowship Award presentation by Aparna, Thursday Nov 17 11-11:30 [www]
  • Technical paper: High-performance Lattice QCD for multi-core based parallel Systems using a cache-friendly hybrid threaded-MPI approach (Intel + Jee) [www]

GT CSE Seminar

I’m giving a talk for the home team today, at Georgia Tech’s School of Computational Science and Engineering (CSE) Seminar. Please drop by to hear about my plans to start a OT500 (Occupy Top 500) movement!

Talk @ CMU

Rich is in drizzly Pittsburgh today visiting a number of great faculty and students at CMU, including members of the amazing SPIRAL team! He’ll also be giving the ECE Seminar a little later today. Can someone please explain the Pittsburgh Left, he asks?

Facing the Multicore Challenge II

Rich will give the lab’s “Balance principles” talk [PDF slides] as an invited keynote at the Facing the Multicore Challenge II conference in Karlsruhe, Germany. His hosts are a delight and the conference, which focuses on young scientists, is energetic and sharp. Overall, a great and highly recommended meeting! [Photo: ZKM Car Culture Exhibit]

Talk @ PPAM’11

I (Rich) am giving a talk today [PDF slides] at the International Conference on Parallel Processing and Applied Mathematics (PPAM), which is being held in the beautiful and historic town of Toru?, Poland. This is a great and well-organized meeting, and I look forward to the next meeting!


Aparna and Rich are in Fort Collins, Colorado today attending the 3rd Annual Concurrent Collections (CnC) Workshop. Ryan Newton (Indiana University) is the lead organizer. Talk slides will be posted at the workshop website shortly. [www]