Conference Program

Wednesday, June 20
08:45 - 09:00Conference Opening
09:00 - 10:00Session 1: Keynote 1
 Chair: Matei Ripeanu
 Putting "Big-data" to Good Use: Building Kinect
Mihai Budiu, Microsoft Research, Mountain View, USA
10:00 - 10:20Break
10:20 - 12:00Session 2: Virtualization
 Chair: Beth Plale
 vSlicer: Latency-aware Virtual Machine Scheduling via Differentiated-frequency CPU Slicing
Cong Xu (Purdue University), Sahan Gamage (Purdue University), Pawan N. Rao (Purdue University), Ardalan Kangarlou (NetApp), Ramana Kompella (Purdue University), Dongyan Xu (Purdue University)
 Singleton: System-wide Page Deduplication in Virtual Environments
Prateek Sharma, Purushottam Kulkarni (IIT Bombay)
 Locality-aware Dynamic VM Reconfiguration on MapReduce Clouds
Jongse Park, Daewoo Lee, Bokyeong Kim, Jaehyuk Huh, Seungryoul Maeng (KAIST)
 Achieving Application-Centric Performance Targets via Consolidation on Multicores: Myth or Reality?
Lydia Y. Chen Chen (IBM Research Zurich Lab), Danilo Ansaloni (University of Lugano), Evgenia Smirni (College of William and Mary), Akira Yokokawa (University of Lugano), Walter Binder (University of Lugano)
12:00 - 13:15Lunch
13:15 - 14:55Session 3: I/O
 Chair: Dongyan Xu
 Enabling Event Tracing at Leadership-Class Scale through I/O Forwarding Middleware
Thomas Ilsche (Technische Universität Dresden), Joseph Schuchart (Technische Universität Dresden), Jason Cope (Argonne National Laboratory), Dries Kimpe (Argonne National Laboratory), Terry Jones (Oak Ridge National Laboratory), Andreas Knöpfer (Technische Universität Dresden), Kamil Iskra (Argonne National Laboratory), Robert Ross (Argonne National Laboratory), Wolfgang E. Nagel (Technische Universität Dresden), Stephen Poole (Oak Ridge National Laboratory)
 ISOBAR Hybrid Compression-I/O Interleaving for Large-scale Parallel I/O Optimization
Eric R. Schendel (North Carolina State University), Saurabh V. Pendse (North Carolina State University), John Jenkins (North Carolina State University), David A. Boyuka (North Carolina State University), Zhenhuan Gong (North Carolina State University), Sriram Lakshminarasimhan (North Carolina State University), Qing Liu (Oak Ridge National Laboratory), Scott Klasky (Oak Ridge National Laboratory), Robert Ross (Argonne National Laboratory), Nagiza F. Samatova (North Carolina State University)
 QBox: Guaranteeing I/O Performance on Black Box Storage Systems
Dimitris Skourtis, Shinpei Kato, Scott Brandt (University of California, Santa Cruz)
 Towards Efficient Live Migration of I/O Intensive Workloads: A Transparent Storage Transfer Proposal
Bogdan Nicolae (INRIA), Franck Cappello (INRIA/UIUC)
14:55 - 15:15Break
15:15 - 16:05Session 4: Industry Session (more info)
 Chair: Alexandru Iosup
 Parallelization of C Programs through Dependency Analysis
Jos van Eijndhoven, CTO of Vector Fabrics, Eindhoven, the Netherlands
 Parallel Programming for the Masses
Vincent Hindriksen, StreamComputing, Haarlem, the Netherlands
16:05 - 16:15Short Break
16:15 - 17:05Session 5: GPUs
 Chair: Ana Varbanescu
 A Virtual Memory Based Runtime to Support Multi-tenancy in Clusters with GPUs
Michela Becchi (University of Missouri), Kittisak Sajjapongse (University of Missouri), Ian Graves (University of Missouri), Adam Procter (University of Missouri), Vignesh Ravi (Ohio State University), Srimat Chakradhar (NEC Laboratories America)
 Interference-driven Scheduling and Resource Management for GPU-based Heterogeneous Clusters
Rajat Phull, Cheng-Hong Li, Kunal Rao, Hari Cadambi, Srimat Chakradhar (NEC Laboratories America)
17:05 - 17:50Session 6: Poster Presentations (more info)
 Chair: Ana Varbanescu
18:00 - 19:00Session 7: Posters + Conference Reception


Thursday, June 21
09:00 - 10:00Session 8: Keynote 2
 Chair: Thilo Kielmann
 Leveraging Renewable Energy in Data Centers: Present and Future
Ricardo Bianchini, Rutgers University, USA
10:00 - 10:20Break
10:20 - 12:00Session 9: Applications and Resources
 Chair: Alexandru Iosup
 Work Stealing and Persistence-based Load Balancers for Iterative Overdecomposed Applications
Jonathan Lifflander (UIUC), Sriram Krishnamoorthy (PNNL), Laxmikant V. Kale (UIUC)
 Highly Scalable Graph Search for the Graph500 Benchmark
Koji Ueno (Tokyo Institute of Technology/JST CREST), Toyotaro Suzumura (Tokyo Institute of Technology/IBM Research Tokyo/JST CREST)
 PonD : Dynamic Creation of HTC Pool on Demand Using a Decentralized Resource Discovery System
Kyungyong Lee (University of Florida), David Wolinsky (Yale University), Renato Figueiredo (University of Florida)
 SpeQuloS: A QoS Service for BoT Applications Using Best Effort Distributed Computing Infrastructures
Simon Delamare (INRIA), Gilles Fedak (INRIA), Derrick Kondo (INRIA), Oleg Lodygensky (IN2P3)
12:00 - 13:15Lunch
13:15 - 14:55Session 10: MapReduce
 Chair: Carlos Maltzahn
 Understanding the Effects and Implications of Compute Node Related Failures in Hadoop
Florin Dinu, T. S. Eugene Ng (Rice University)
 Optimizing MapReduce for GPUs with Effective Shared Memory Usage
Linchuan Chen, Gagan Agrawal (The Ohio State University)
 CAM: A Topology Aware Minimum Cost Flow Based Resource Manager for MapReduce Applications in the Cloud
Min Li (Virginia Tech), Dinesh Subhraveti (IBM Almaden Research Center), Ali Butt (Virginia Tech), Aleksandr Khasymski (Virginia Tech), Prasenjit Sarkar (IBM Almaden Research Center)
 Distributed Approximate Spectral Clustering for Large-Scale Datasets
Fei Gao (Simon Fraser University), Wael Abd-Almageed (University of Maryland), Mohamed Hefeeda (Qatar Computing Research Institute)
14:55 - 15:15Break
15:15 - 16:05Session 11: Energy
 Chair: Henri Bal
 Exploring Cross-layer Power Management for PGAS Applications on the SCC Platform
Marc Gamell (Rutgers University), Ivan Rodero (Rutgers University), Manish Parashar (Rutgers University), Rajeev Muralidhar (Intel India)
 Dynamic Adaptive Virtual Core Mapping to Improve Power, Energy, and Performance in Multi-socket Multicores
Chang Bae (Northwestern University), Lei Xia (Northwestern University), Peter Dinda (Northwestern University), John Lange (University of Pittsburgh)
16:05 - 16:15Short Break
16:15 - 17:15Session 12: Panel on Energy Efficiency
 Moderator: Matei Ripeanu, The University of British Columbia, Vancouver, Canada
Ricardo Bianchini (Rutgers University)
Barney Maccabe (Oak Ridge National Laboratory)
Manish Parashar (Rutgers University)
Karsten Schwan (Georgia Tech)
18:00 - 19:00Social Event (1): Visit to Museum "De Prinsenhof"
19:00 - 22:30Social Event (2): Conference Dinner in Restaurant "De Prinsenkelder"


Friday, June 22
09:00 - 10:00Session 13: Achievement Award Talk
 Chair: Dick Epema
 Reflections on 20 Years of Grid Computing
Ian Foster, University of Chicago and Argonne National Laboratory, USA
10:00 - 10:20Break
10:20 - 11:35Session 14: Networked Systems
 Chair: Renato Figueiredo
 VNET/P: Bridging the Cloud and High Performance Computing Through Fast Overlay Networking
Lei Xia (Northwestern University), Zheng Cui (University of New Mexico), John Lange (University of Pittsburgh), Yuan Tang (UESTC, China), Peter Dinda (Northwestern University), Patrick Bridges (University of New Mexico)
 Massively-Parallel Stream Processing under QoS Constraints with Nephele
Björn Lohrmann, Daniel Warneke, Odej Kao (Technische Universität Berlin)
 A Resiliency Model for High Performance Infrastructure Based on Logical Encapsulation
James Moore (The University of Southern California/EMC Corporation), Carl Kesselman (The University of Southern California)
11:35 - 11:45Short Break
11:45 - 12:45Session 15: Panel on the Value of HPDC
 Moderator: Henri Bal, Vrije Universiteit, The Netherlands
Renato Figueiredo (University of Florida)
Ian Foster (University of Chicago and Argonne National Laboratory)
Rob van Nieuwpoort (Netherlands eScience Center)
Jon Weissmann (University of Minnesota)
12:45 - 13:00Conference Closing (with Best Paper Award)


Industry Session Abstracts and Bios

Jos van Eijndhoven, CTO of Vector Fabrics, Eindhoven, the Netherlands
Title: Parallelization of C Programs through Dependency Analysis

Abstract: Most C(++) application developers still create sequential programs, focusing on correct functionality. Parallelization is a subsequent optimization step to obtain satisfactory execution times on modern concurrent computer hardware. Unfortunately, parallelizing sequential C programs might introduce hard to find bugs, and is a cumbersome task for larger programs. Dynamic data-dependency analysis can reveal parallelization opportunities and avoid implementation errors. The tractability of this approach will be shown for the Lammps molecular simulator HPC application.

Bio: Jos van Eijndhoven is CTO and co-founder of Vector Fabrics BV, a company that creates and markets tools for mapping applications to multi-core architectures. Earlier, he was principle architect at Philips/NXP Research, involved in media-processor architecture and their application programming. Jos received a PhD from Eindhoven University of Technology, the Netherlands. He holds 15 worldwide patents and is (co-)author of about 100 scientific publications.

Vincent Hindriksen, StreamComputing, Haarlem, the Netherlands
Title: Parallel Programming for the Masses

Abstract: During the two years since StreamComputing's founding, I have been in a lot of discussions on the best way of parallelizing loops, from using pfor and data-oriented kernels to automatic unrolling and very intelligent compilers. I want to discuss choices developers and architects have to solve the parallelization problem, and moreover, how to cope with decades of legacy. As there is not one right answer, I want to give a complete-as-possible overview of what we want versus what we need. This includes current buzz-words like LLVM, CUDA and OpenCL, and an introduction to the multi-hardware-architecture and multi-OS/platform era we are entering.

Bio: Vincent Hindriksen founded StreamComputing two years ago and has since then focused on speeding up computations in software using techniques that work on more processors than just CPUs. The focus is on hardware architectures, OpenCL and since this year also LLVM. The years before he sped up the office software of his project outside the official working hours. For the years to come he wants to keep helping companies prepare for an era where there is not only one processor, the X86 by Intel/AMD, but many more.


Accepted Posters (based on full submitted papers)

  • A1. Dynamic Binary Rewriting and Migration for Shared-ISA Asymmetric, Multicore Processors
    Giorgis Georgakoudis (University of Thessaly), Dimitrios S. Nikolopoulos (Queen's University of Belfast)
  • A2. Exploring the Performance and Mapping of HPC Applications to Platforms in the Cloud
    Abhishek Gupta (UIUC), Laxmikant V. Kale (UIUC), Dejan S. Milojicic (HP labs, Palo Alto), Paolo Faraboschi (HP labs, Palo Alto), Richard Kaufmann (HP labs, Palo Alto), Verdi March (HP labs, Singapore), Filippo Gioachin (HP labs, Singapore), Chun Hui Suen (HP labs, Singapore), Bu-Sung Lee (HP labs, Singapore)
  • A3. Fault Tolerant Data Intensive Algorithms
    Mucahid Kutlu, Gagan Agrawal, Oguz Kurt (The Ohio State University)
  • A4. P*: A Model of Pilot-Abstractions
    Andre Luckow (CCT/LSU), Mark Santcroos (AMC/University of Amsterdam), Ole Weidner (CCT/LSU), Andre Merzky (CCT/LSU), Sharath Maddineni (CCT/LSU), Shantenu Jha (Rutgers University)
  • A5. Performance Evaluation of Inter-thread Communication Mechanisms on Multicore/multithreaded Architectures
    Davide Pasetto, Massimiliano Meneghin, Hubertus Franke, Fabrizio Petrini, Jimi Xenidis (IBM Research)
  • A6. SMART-IO: SysteM-AwaRe Two-Level Data Organization for Efficient Scientific Analytics
    Yuan Tian (Auburn University), Scott Klasky (Oak Ridge National Laboratories), Weikuan Yu (Auburn University), Hasan Abbasi (Oak Ridge National Laboratories), Bin Wang (Auburn University), Norbert Podhorszki (Oak Ridge National Laboratories), Ray Grout (National Renewable Energy Laboratory), Matt Wolf (Georgia Institute of Technology)
  • A7. Coupling Task Progress for MapReduce Resource-Aware Scheduling
    Jian Tan, Xiaoqiao Meng, Li Zhang (IBM T.J. Watson)


Open Call Posters (based on the call for posters)