April 6-7, 2016 - X-Stack PI Meeting: Difference between revisions
From Modelado Foundation
imported>Achien No edit summary |
imported>Jayaajay (Adding XPRESS Demos) |
||
| Line 43: | Line 43: | ||
* 8. [https://youtu.be/lt_3RPz6FAc DEGAS: HipMer - A High performance distributed memory scalable version of Meraculous] | * 8. [https://youtu.be/lt_3RPz6FAc DEGAS: HipMer - A High performance distributed memory scalable version of Meraculous] | ||
* 9. [https://drive.google.com/a/cas.uoregon.edu/file/d/0B-I2mdLV09Y0QnNFQ1o4OUNKNTA/view?pref=2&pli=1 D-TEC: A CAFe (DSL) transformed program running on Titan] | * 9. [https://drive.google.com/a/cas.uoregon.edu/file/d/0B-I2mdLV09Y0QnNFQ1o4OUNKNTA/view?pref=2&pli=1 D-TEC: A CAFe (DSL) transformed program running on Titan] | ||
*10.[https://youtu.be/VdBgHlK083A GVR (D-TEC): Version-based Resilience: Performance at Scale, Flexible Recovery, and Fast Versioning] and Demo Slides [[Media:XStack16_GVR_demo.pptx]] | *10. [https://youtu.be/VdBgHlK083A GVR (D-TEC): Version-based Resilience: Performance at Scale, Flexible Recovery, and Fast Versioning] and Demo Slides [[Media:XStack16_GVR_demo.pptx]] | ||
* 11. [https://iu.box.com/s/uxc2bhwj88bmoeogan92u0fnqi2llgbm XPRESS: HPX-5 Running on Cori@NERSC (Cray XC-40) with dynamic scheduling to stay under soft power cap] | |||
* 12. [https://iu.app.box.com/s/jvxuihfsecdqfygnatdbzkk2rcub6593 XPRESS: Automatic, Dynamic Load Balancing of FMM in HPX-5] | |||
* 13. [https://iu.box.com/s/nu2xe4x102z6nme8q394gwt8jevfpkd4 XPRESS: Dynamic, Adaptive Execution of LULESH in HPX-5] | |||
* 14. [https://drive.google.com/file/d/0By-HAV_CacKcSkVCYnpFR2hvVzA/view?usp=drive_web XPRESS: HPXCL demonstration] | |||
== Breakout Sessions == | == Breakout Sessions == | ||
Revision as of 20:58, April 29, 2016
Meeting Information
When: April 6-7, 2016
Where: Lawrence Berkeley National Laboratory
Meeting Agenda
Meeting Presentations April 6
- 1. X-Stack PI Meeting Welcome and Goals (Vivek Sarkar and Sonia Sachs)
- 2. DEGAS: Dynamic Exascale Global Address Space (Yili Zheng)
- 3. The D-TEC Project (Dan Quinlan and Saman Amarasinghe)
- 4. Traleika Glacier (Josh Fryman)
- 5. XPRESS Project (Ron Brightwell)
- 6. PIPER: Performance Insights for Programmers and Exascale Runtimes (Todd Gamblin)
- 7. SLEEC (Milind Kulkarni)
- 8. Corvette (Koushik Sen)
- 9. X-TUNE (Mary Hall)
- 10. Vancouver (Jeffrey Vetter)
- 11. ARES (Jeffrey Vetter)
Meeting Presentations April 7
- 1. Next Generation AMR: An Application Developer’s Perspective (Ann Almgren)
- 2. Future Programming Challenges for DOE from the Applica/on Developer Perspective (David Richards)
- 3. Tools for Exascale Software Development from the Application Perspective: The Importance of DSL Technology (Paul Woodward)
- 4. Distributed Learning Dynamics Convergence in Routing Games (Alexander Bayen)
Technology Demonstrations
- 1. D-TEC: Halide DSL for stencil computations video
- 2. D-TEC: Using the ROSE framework for code generation and optimizations
- 3. DEGAS: HPCToolkit
- 4. PIPER: Measuring and Attributing Performance of Applications that Employ Emerging Template-based Parallel Programming Models
- 5. PIPER: Integration of Active Harmony and Caliper for Improved Tuning Accuracy
- 6. DEGAS: End-to-End Application Resilience to Protect Against Soft Errors
- 7. DEGAS: PGAS Containment Domains Runtime Demo
- 8. DEGAS: HipMer - A High performance distributed memory scalable version of Meraculous
- 9. D-TEC: A CAFe (DSL) transformed program running on Titan
- 10. GVR (D-TEC): Version-based Resilience: Performance at Scale, Flexible Recovery, and Fast Versioning and Demo Slides Media:XStack16_GVR_demo.pptx
- 11. XPRESS: HPX-5 Running on Cori@NERSC (Cray XC-40) with dynamic scheduling to stay under soft power cap
- 12. XPRESS: Automatic, Dynamic Load Balancing of FMM in HPX-5
- 13. XPRESS: Dynamic, Adaptive Execution of LULESH in HPX-5
- 14. XPRESS: HPXCL demonstration