Keynote: "Accelerating Datacenter Services with Reconfigurable Logic”
Aaron Smith, Microsoft Research, USA (Details).
All papers are now available in arXiv.
We accepted 7 papers out of 12.
Each paper had at least 3 reviews.
| 10:00-10:05 |
Workshop introduction |
| 10:05-10:30 |
Automatic Tuning of Local Memory Use on GPGPUs
Tianyi David Han and Tarek Abdelrahman University of Toronto, Canada |
| 10:30-10:55 |
Adapting Graph Application Performance Via Alternate Data Structure Representations
Amlan Kusum, Iulian Neamtiu and Rajiv Gupta University Of California, Riverside, USA |
| 11:00-11:30 | Break |
| 11:30-11:55 |
Easy-to-Use On-the-Fly Binary Program Acceleration on Many-Cores Marvin Damschen and Christian Plessl University of Paderborn, Germany |
| 11:55-12:20 |
Performance comparison between Java and JNI for optimal implementation of computational micro-kernels Nassim Halli1 , Henri-Pierre Charles2 and Jean-François Mehaut3 1 Aselta-nanographics, France 2 CEA-LIST, France 3 Université de Grenoble, France |
| 12:20-12:45 |
Self-Repairing Disk Arrays Jehan-Francois Paris1, Ahmed Amer2, Darrell Long3 and Thomas Schwarz4 1 University of Houston, USA 2 Santa Clara University, USA 3 University of California, Santa Cruz 4 Universidad Catolica Del Uruguay, Uruguay This paper has an interesting discussion on Slashdot (supporting our new publication model proposal): [Slashdot Link] |
| 12:45-14:15 | Lunch |
| 14:15-15:30 |
Keynote: "Accelerating Datacenter Services with Reconfigurable Logic” Aaron Smith (Microsoft Research, USA) |
| 15:30-16:00 | Break |
| 16:00-16:25 |
Adaptive and application dependent runtime guided hardware prefetcher reconfiguration on the IBM POWER7 David Prat Robles, Marc Casas Guix, Miquel Moretó Planas and Mateo Valero I Cortés Barcelona Supercomputing Center (BSC), Spain |
| 16:25-16:50 |
A Self-adaptive Auto-scaling Method for Scientific Applications on HPC Environments and Clouds
(Nvidia best paper award)
Kiran Mantripragada1, Alecio Binotto1 and Leonardo Tizzei2 1 IBM Research - Brazil 2 IBM Brazil |
| 16:50-17:00 | NVIDIA best paper award ceremony and workshop conclusions |
| 17:00-17:30 |
Common discussion on artifact evaluation experience from CGO'15/PPoPP'15 and community-driven reviewing
Grigori Fursin (cTuning Foundation, France) |
