News and Announcements

Summit

Even as Titan, ORNL’s Cray XK7 supercomputer and current No. 2 on the TOP500, keeps churning away at ~18 PetaFLOP/s, the DOE and the Oak Ridge Leadership Computing Facility (OLCF) are making plans for the next push towards Exascale. This push comes from a new machine, Summit, which is being designed and built through a partnership between NVIDIA, IBM, and Mellanox, which will provide the accelerators, CPUs, and interconnect, respectively.

Summit will arrive sometime in 2018 as part of a three-pronged attack by the DOE called the Collaboration of Oak Ridge, Argonne, and Livermore (CORAL). As the name suggests, ORNL, ANL, and LLNL will all host new supercomputers, each striving towards 150 PetaFLOP/s with a possible expansion to 300 PetaFLOP/s or more.

Things are about to get interesting, especially since the Chinese have announced their five year plan to achieve Exascale (1000 PetaFLOP/s) in 2020. Stay tuned.

Conference Reports

ASC16 Student Supercomputer Challenge

  • Jack_asc
  • asc4
  • asc3
  • asc2
  • Jack_asc

 

ICL’s Jack Dongarra served as co-chair of the Expert Committee and judge of the Asia Supercomputing Community’s 2016 Student Supercomputing Challenge (ASC16). The challenge, hosted by the Huazhong University of Science and Technology on April 18-22, 2016, is billed as the world’s largest supercomputing hackathon.

For 2016, 16 finalists were selected from 175 registered teams worldwide after two months of rigorous evaluation in the preliminary contest. For the final round, the finalists designed and built a cluster to run HPL, HPCG, surface wave numerical model MASNUM, material simulation software ABINIT, deep learning algorithm DNN, and a Mystery Application under 3000W. The Inspur Group provided all necessary hardware.

Each team presented system designs, the final result, and analysis to the Judge Committee, of which Jack was a member. The team from Huazhong University of Science and Technology (home team) won the day and were crowned champions of ASC16. A full listing of the results is available here.

GPU Technology Conference

26150345542_368c7296c1_k

On April 4th, 2016, ICL sent a small contingent of the MAGMA research and development team to NVIDIA’s GPU Technology Conference (GTC 2016) in San Jose, CA. GTC 2016 is billed as the largest and most important event of the year for GPU developers, and showcases the bleeding edge applications that utilize GPU technology, including artificial intelligence and deep learning, virtual reality, and self-driving cars.

Stan Tomov and Azzam Haidar co-presented a talk/tutorial on High-Performance Batched Computations for GPUs: Approaches and Applications, outlining techniques for efficient batched computations on GPUs, where small and independent computations must be grouped and executed together to obtain high performance.

Ahmad Abdelfattah presented a poster describing Cholesky Factorization on Batches of Matrices with Fixed and Variable Sizes, wherein he describes a high performance solution for Cholesky factorization on batches of relatively small matrices; the proposed solution outperforms most of the existing state-of-the-art techniques for solving batched problems.

Interview

Cindy Knisley Then

Cindy Knisley

Where are you from, originally?

I was born right here in Knoxville (at the old St. Mary’s Hospital). I grew up out in the county—in the Corryton community.

Can you summarize your educational background?

I went to Gibbs High School in Corryton, where I had the second highest GPA in my class. After high school, I went to Walters State Community College in Morristown, Tennessee. I graduated Summa Cum Laude with an Associate’s degree in Accounting. Six years later, I went back to school at night and graduated Magna Cum Laude from Tusculum College with a degree in Business Management.

Where did you work before joining ICL?

I started my career as an assistant controller with a local wholesale building supply company. After being there three years, I took a job at Kimberly-Clark Corporation. I worked there for ten years, then I stayed home for about eight years when we lived in Middle Tennessee. I worked a couple of little jobs to get me back in the workforce before coming to UT in 2011.

How did you first hear about the lab and what made you want to work here?

I first heard about the lab from the UT recruiting site. I had been working in the Biology Business Office and was looking for more of a challenge.

What is your primary role here at ICL? What are you working on?

I am, of course, the Financial Specialist at ICL. What that means is that I make sure we spend our grant and gift money wisely and appropriately. We have different rules from each funding agency, as well as the ever-changing policies and procedures set forth by UT. I may seem like the policy patrol, but that is why I am here! I try my best to keep us on the straight and narrow.

What are your interests/hobbies outside of work?

For years, my interests and hobbies all revolved around my three children. Since we are empty-nesters now, I am trying to discover what I enjoy doing. I have always loved the water. Two of the things I enjoy most, but rarely get to do, are swimming and fishing. When we lived in Middle Tennessee, we were surrounded by Kentucky Lake. We lived two miles from Pebble Isle Marina where we kept a pontoon boat. We were also near rivers and creeks. I loved it!

I also like the idea of making things myself, like crocheting and sewing. I built a “doggie condo” last fall with insulation, shingled roof, partitioned room, and a covered deck. It doesn’t look so great, but I really enjoyed building it.

Tell us something about yourself that might surprise people.

You can’t tell by looking at me now, but I was an athlete. I played softball from 5 years old, and was a pitcher in my later years. I was pretty good, even if I do say so myself. I played basketball until the 8th grade, when I stopped growing. (Go ahead and laugh!). After that, I was a cheerleader the next four years. But, probably the most surprising is that I can beat any one of you in a watermelon seed spitting contest! I learned when my siblings and I would have seed spitting wars, where if your seed stuck on someone else’s face, they were out of the game. Yes, I am from the country!

If you weren’t working at ICL, where would you like to be working and why?

If I were not working at ICL, I would like to be working anywhere with an office overlooking the ocean (and accessible to the beach and swimming pool).

Recent Papers

  1. Valero-Lara, P., J. Dongarra, A. Haidar, S. D. Relton, S. Tomov, and M. Zounon, A Standard for Batched BLAS Routines , Paris, France, 17th SIAM Conference on Parallel Processing for Scientific Computing (SIAM PP16), April 2016.  (1.93 MB)
  2. Abdelfattah, A., A. Haidar, S. Tomov, and J. Dongarra, Cholesky Factorization on Batches of Matrices with Fixed and Variable Sizes , San Jose, CA, GPU Technology Conference (GTC16), Poster, April 2016.  (480.51 KB)
  3. Anzt, H., J. Dongarra, M. Kreutzer, G. Wellein, and M. Kohler, Efficiency of General Krylov Methods on GPUs – An Experimental Study,” 2016 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pp. 683-691, May 2016. DOI: 10.1109/IPDPSW.2016.45
  4. Anzt, H., J. Dongarra, M. Kreutzer, G. Wellein, and M. Kohler, Efficiency of General Krylov Methods on GPUs – An Experimental Study,” The Sixth International Workshop on Accelerators and Hybrid Exascale Systems (AsHES), Chicago, IL, IEEE, May 2016. DOI: 10.1109/IPDPSW.2016.45  (285.28 KB)
  5. Jia, Y., P. Luszczek, and J. Dongarra, Hessenberg Reduction with Transient Error Resilience on GPU-Based Hybrid Architectures,” 30th IEEE International Parallel & Distributed Processing Symposium (IPDPS), Chicago, IL, IEEE, May 2016.  (535.72 KB)
  6. Newburn, C. J., G. Bansal, M. Wood, L. Crivelli, J. Planas, A. Duran, P. Souza, L. Borges, P. Luszczek, S. Tomov, et al., Heterogeneous Streaming,” The Sixth International Workshop on Accelerators and Hybrid Exascale Systems (AsHES), IPDPS 2016, Chicago, IL, IEEE, May 2016.  (2.73 MB)
  7. Abdelfattah, A., H. Anzt, J. Dongarra, M. Gates, A. Haidar, J. Kurzak, P. Luszczek, S. Tomov, , and A. YarKhan, Linear Algebra Software for Large-Scale Accelerated Multicore Computing,” Acta Numerica, vol. 25, pp. 1-160, May 2016. DOI: 10.1017/S0962492916000015
  8. Abdelfattah, A., A. Haidar, S. Tomov, and J. Dongarra, On the Development of Variable Size Batched Computation for Heterogeneous Parallel Architectures,” The 17th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC 2016), IPDPS 2016, Chicago, IL, IEEE, May 2016.  (708.62 KB)
  9. Benoit, A., A. Cavelan, Y. Robert, and H. Sun, Optimal Resilience Patterns to Cope with Fail-stop and Silent Errors,” 2016 IEEE International Parallel and Distributed Processing Symposium (IPDPS), Chicago, IL, IEEE, May 2016. DOI: 10.1109/IPDPS.2016.39  (603.58 KB)
  10. Haugen, B., Performance Analysis and Modeling of Task-Based Runtimes,” Department of Electrical Engineering and Computer Science, vol. PhD, Knoxville, University of Tennessee, May 2016.  (5.14 MB)
  11. Abdelfattah, A., H. Ltaeif, D. Keyes, and J. Dongarra, Performance optimization of Sparse Matrix-Vector Multiplication for multi-component PDE-based applications using GPUs,” Concurrency and Computation: Practice and Experience, vol. 28, issue 12, pp. 3447 - 3465, May 2016. DOI: 10.1002/cpe.v28.1210.1002/cpe.3874  (3.21 MB)
  12. Luszczek, P., M. Gates, J. Kurzak, A. Danalis, and J. Dongarra, Search Space Generation and Pruning System for Autotuners,” 30th IEEE International Parallel & Distributed Processing Symposium (IPDPS), Chicago, IL, IEEE, May 2016.  (555.44 KB)

Recent Conferences

  1. APR
    Ahmad Abdelfattah Ahmad
    Ahmad
    Azzam Haidar
    Azzam
    Stanimire Tomov
    Stan
    Ahmad Abdelfattah Ahmad, Azzam Haidar, Stanimire Tomov
  2. APR
    -
    SIAM PP Paris, France
    Aurelien Bouteiller
    Aurelien
    George Bosilca
    George
    Ichitaro Yamazaki
    Ichitaro
    Aurelien Bouteiller, George Bosilca, Ichitaro Yamazaki
  3. MAY
    Piotr Luszczek
    Piotr
    Piotr Luszczek
  4. MAY
    TESSE Meeting New York, New York
    Amina Guermouche
    Amina
    George Bosilca
    George
    Thomas Herault
    Thomas
    Amina Guermouche, George Bosilca, Thomas Herault
  5. MAY
    Amina Guermouche
    Amina
    Aurelien Bouteiller
    Aurelien
    George Bosilca
    George
    Ichitaro Yamazaki
    Ichitaro
    Thomas Herault
    Thomas
    Amina Guermouche, Aurelien Bouteiller, George Bosilca, Ichitaro Yamazaki, Thomas Herault
  6. MAY
    -
    Ahmad Ahmad
    Ahmad
    Azzam Haidar
    Azzam
    Hartwig Anzt
    Hartwig
    Jack Dongarra
    Jack
    Jakub Kurzak
    Jakub
    Mark Gates
    Mark
    Piotr Luszczek
    Piotr
    Stanimire Tomov
    Stan
    Terry Moore
    Terry
    Ahmad Ahmad, Azzam Haidar, Hartwig Anzt, Jack Dongarra, Jakub Kurzak, Mark Gates, Piotr Luszczek, Stanimire Tomov, Terry Moore
  7. MAY
    George Bosilca
    George
    Piotr Luszczek
    Piotr
    George Bosilca, Piotr Luszczek
  8. MAY
    Ahmad Abdelfattah Ahmad
    Ahmad
    Ahmad Abdelfattah Ahmad

Upcoming Conferences

  1. JUN
    Heike Jagode
    Heike
    Thomas Herault
    Thomas
    Heike Jagode, Thomas Herault
  2. JUN
    Stanimire Tomov
    Stan
    Stanimire Tomov
  3. JUN
    MPI Forum WA Bellevue, Washington
    Aurelien Bouteiller
    Aurelien
    Aurelien Bouteiller
  4. JUN
    ADAC Workshop Luguano, Switzerland
    Azzam Haidar
    Azzam
    Jack Dongarra
    Jack
    Jakub Kurzak
    Jakub
    Azzam Haidar, Jack Dongarra, Jakub Kurzak
  5. JUN
    -
    Jack Dongarra
    Jack
    Terry Moore
    Terry
    Tracy Rafferty
    Tracy
    Jack Dongarra, Terry Moore, Tracy Rafferty
  6. JUN
    -
    Azzam Haidar
    Azzam
    Jack Dongarra
    Jack
    Jakub Kurzak
    Jakub
    Terry Moore
    Terry
    Azzam Haidar, Jack Dongarra, Jakub Kurzak, Terry Moore

Recent Lunch Talks

  1. APR
    1
    Ahmad Ahmad
    Ahmad Ahmad
    On the Development of Variable-Size Batched Computation for Heterogeneous Parallel Architectures PDF
  2. APR
    8
    Piotr Luszczek
    Piotr Luszczek
    Search Space Description, Generation, and Pruning System for Autotuners PDF
  3. APR
    15
    Miro Stoyanov
    Miro Stoyanov
    ORNL
    Resilient Solvers for Partial Differential Equations PDF
  4. APR
    22
    Chongxiao Cao
    Chongxiao Cao
    Fault Tolerant Design for a Task-based Runtime PDF
  5. APR
    29
    Wei Wu
    Wei Wu
    Accelerator Integration with Programming Models PDF
  6. MAY
    4
    Yaohung Tsai
    Yaohung Tsai
    AlphaGo: The Go AI from Google DeepMind
  7. MAY
    6
    Oleg Shylo
    Oleg Shylo
    Department of Industrial & Systems Engineering at UTK
    Scalable Communication for Parallel Optimization
  8. MAY
    13
    George Bosilca
    George Bosilca
    PaRSEC - Yet another runtime? PDF
  9. MAY
    20
    Iain Duff
    Iain Duff
    the Numerical Analysis Group at the Scientific Computing Department of the Science and Technology Facilities Council (UK)
    Scalability of Sparse Direct Codes PDF
  10. MAY
    27
    Azzam Haidar
    Azzam Haidar
    Heterogeneous Computation: The Current Challenge

Upcoming Lunch Talks

  1. JUN
    10
    Emmanuel Agullo
    Emmanuel Agullo
    INRIA
    Overview of Task-based Sparse and Data-sparse Solvers on Top of Runtime Systems
  2. JUN
    17
    Julien Langou
    Julien Langou
    University of Colorado
    A Makespan Lower Bound for the Scheduling of the Tiled Cholesky Factorization based on ALAP scheduling PDF

congratulations

Zachary Nicholas Dongarra

Nick and Whitney Dongarra welcomed their new baby boy, Zachary, to the world on April 5, 2016. Both baby and mother are doing well. Grandpa Jack is pictured below with his (now) two grand children, Zachary and Rebecca. Congratulations to the Dongarra family!

grandpa_jack

Dates to Remember

2016 ICL Retreat

The 2016 ICL retreat has been set for August 15th – 16th at the RT Lodge in Maryville, TN. Mark your calendars.