Clusters, Clouds, and Data for Scientific Computing

CCDSC 2012

September 11^th – 14^th, 2012

Châteauform’

La Maison des Contes

427 Chemin de Chanzé, France

Sponsored by:

Clusters, Clouds, and Data for Scientific Computing

 2012

Châteauform’

La Maison des Contes

427 Chemin de Chanzé, France

September 11^th – 14^th, 2012

CCDSC 2012 will be held at a resort outside of Lyon France called La maison des contes http://www.chateauform.com/en/chateauform/maison/17/chateau-la-maison-des-contes

The address of the Chateau is:

Châteauform’ La Maison des Contes

427 chemin de Chanzé

69490 Dareizé

Telephone: +33 1 30 28 69 69

1 hr 30 min from the Saint Exupéry Airport

45 minutes from Lyon

GPS Coordinates: North latitude 45° 54' 20" East longitude 4° 30' 41"

Go to http://maps.google.com and type in: “427 chemin de Chanzé 69490 Dareizé” or see:

Maps: http://www.cs.utk.edu/~dongarra/CCGSC-2012/maps.pdf

Map of Chateau: http://www.cs.utk.edu/~dongarra/CCGSC-2012/chateau.pdf

Message from the Program Chairs

This proceeding gathers information about the participants of the Workshop on Clusters, Clouds, and Data for Scientific Computing that will be held at La Maison des Contes, 427 Chemin de Chanzé, France on September 11th – 14th, 2012.  This workshop is a continuation of a series of workshops started in 1992 entitled Workshop on Environments and Tools for Parallel Scientific Computing. These workshops have been held every two years and alternate between the U.S. and France. The purpose of this the workshop, which is by invitation only, is to evaluate the state-of-the-art and future trends for cluster computing and the use of computational clouds for scientific computing.

This workshop addresses a number of themes for developing and using both cluster and computational clouds. In particular, the talks covered:

§ Survey and analyze the key deployment, operational and usage issues for clusters, clouds and grids, especially focusing on discontinuities produced by multicore and hybrid architectures, data intensive science, and the increasing need for wide area/local area interaction.

§ Document the current state-of-the-art in each of these areas, identifying interesting questions and limitations. Experiences with clusters, clouds and grids relative to their science research communities and science domains that are benefitting from the technology.

§ Explore interoperability among disparate clouds as well as interoperability between various clouds and grids and the impact on the domain sciences.

§ Explore directions for future research and development against the background of disruptive trends and technologies and the recognized gaps in the current state-of-the-art.

Speakers will present their research and interact with all the participants on the future software technologies that will provide for easier use of parallel computers. 

This workshop was made possible thanks to sponsorship from NSF, AMD, ANR, Aristote, Bull, CEA, Cray, France Grilles, Google, Hewlett-Packard, Microsoft Research, The Portland Group, Rhone-Alpes Region with the scientific support of the Innovative Computing Laboratory at the University of Tennessee in Knoxville (UTK) and University Joseph Fourier of Grenoble.

 Thanks!

Jack Dongarra, Knoxville, Tennessee, USA.

Bernard Tourancheau, Grenoble, France

Jack's pictures.

Thilo’s pictures.

Anthony’s pictures.

Draft agenda (9/24/12 11:21 AM)

September 11^th – 14^th, 2012

Tuesday September 11th	Introduction and Welcome               Jack Dongarra, U of Tennessee Bernard Tourancheau, U Grenoble
6:30  – 7:45	Session Chair: Bernard Tourancheau, U Grenoble	(3 talks - 20 minute each)
6:30	Doug Miles, PGI	The Challenges of Portable Heterogeneous Programming
6:50	Luiz DeRose, Cray	Productive Performance on Accelerator Based Systems Using OpenACC Compilers and Tools
7:10	Patrick Demichel, HP	HP-labs researches for Exascale
8:00 pm – 9:00 pm	Dinner
9:00 pm -

Wednesday,               September 12th
7:30 - 8:30	Breakfast
8:30 - 10:35	Session Chair: Jack Dongarra, UTK/ORNL	(5 talks – 25 minutes each)
8:30	Franck Cappello, Univ Paris	Failure prediction: current situation and open questions
8:55	Al Geist, ORNL	Distributed Peer-to-Peer Control in the Cloud
9:20	Thilo Kielmann, Vrije Universiteit	Task Farming in Cloud Environments under Budget Constraints
9:45	George Bosilca, UTK	PaRSEC: A Modular Framework for Performance and Productivity on Hybrid Architectures
10:10	Christian Obrecht, INSA Lyon	Efficient GPU cluster implementation of the lattice Boltzmann method
10:35 -11:00	Coffee
11:00  - 1:05	Session Chair: Laurent Lefevre, INRIA ENS	(5 talks – 25 minutes each)
11:00	Mike Norman, UCSD	Early experiences with Gordon – a flash-based cluster for data-intensive applications
11:25	Fran Berman, RPI	The Digital Society
11:50	Joel Saltz, Emory U	Toward Derivation, Management, and Analysis of Exascale Feature Sets
12:15	Jelena Pjesivac-grbovic, Google	Google Compute Engine: Google solution for large scale data processing and analytics
1:05  - 2:00	Lunch
2:30 – 3:00	Coffee
	Cherri Pancake	ACM SIGHPC
	Al’s Song & Whiskey Presentation
3:00 - 5:30	Panel Chair: Bill Gropp, UIUC
	Rusty Lusk, Argonne	Panel on Crash and Burn: What will succeed and what will fail in Clusters, Clouds, and Data
	Bill Gropp, UIUC
	Satoshi Matsuoka, Tokyo Institute of Technology
	Bernd Mohr, Juelich
	Pete Beckman, ANL
5:45 – 7:30	Wine tasting Cellar Bruno	Travel time to Cellar Bruno 15-20 minutes: two possibilities, walking or cycling
8:00 – 9:00	Dinner
9:00 pm -

Thursday,           September 13th
7:30 - 8:30	Breakfast
8:30 - 10:35	Session Chair: Jean-Yves Berthou, ANR	(4 talks – 25 minutes each)
8:30	Cherri Pancake, Oregon State	Cloud-Enabled Collaboration:  Can We Learn from Past Mistakes?
8:55	Frederic Suter, CNRS/IN2P3	Paving the Road for the Simulation of Exascale and Cloud Systems.
9:20	Geoffrey Fox, Indiana	Robust Parallel Clustering Algorithms
9:45	Laurent Lefevre, INRIA ENS	Energy efficiency in HPC and exascale
10:10	Torsten Hoefler, ETH	Topology Mapping for Petascale++ Supercomputers
10:35 -11:00	Coffee
11:00  - 1:05	Session Chair: Xavier Vigouroux, Bull	(5 talks – 25 minutes each)
11:00	Jeffery Vetter, ORNL & GATech	Aspen: a performance modeling language for codesigning Exascale applications and architectures
11:25	Anne Trefethen, Oxford	Enabling Science using the SKA
11:50	Vaidy Sunderam, Emory U	Towards Computing as a Utility via Adaptive Middleware
12:15	Martin Swany, Indiana U	High Performance Data Transfer with RDMA in the Wide Area
12:40	Jeff Hollingsworth, U Maryland	Automatically Adapting Programs for Mixed-Precision Floating-Point Computation
1:05  - 2:00	Lunch
2:00 – 4:00	Free time
4:00 – 5:00	Coffee
5:00 - 7:05	Session Chair: Frederic Desprez, INRIA	(5 talks – 25 minutes each)
5:00	Anthony Danalis, UTK	Task data flow analysis and extraction from serial input code
5:25	David Walker, Cardiff U	Performance at the Exascale: a Simple Model for Heterogeneous Platforms
5:50	Andrew Lumsdaine, Indiana U	Scalable Graph Algorithms Based on Active Messages
6:15	Andrew Grimshaw, U Virginia	The Global Federated File System: Early Results
6:40	Phil  Papadopoulos, UC San Diego	International Cloud Interoperation Experiments in PRAGMA
8:00 – 9:00	Dinner
9:00 pm -

Friday,  September 14th
7:30 - 8:30	Breakfast
8:30 - 10:35	Session Chair:	(5 talks – 25 minutes each)
8:30	Rajeev Thakur, Argonne	Challenges for Communication Libraries and Runtime Systems at Exascale
8:55	Padma Raghavan, Penn State	Do Fixed-Problem Speedups Matter?
9:20	Emmanuel Jeannot, INRIA	Process placement of MPI an multi-threaded applications. Why, how and for what gain?
9:45	Raymond Namyst, U Bordeaux & INRIA	Programming heterogeneous, accelerator-based multicore machines: a runtime system's perspective
10:10	Ewa Deelman, ISI	Hosted Science: How to support complex scientific applications on the cloud
10:35 -11:00	Coffee
11:00  - 1:05	Session Chair:	(4 talks – 25 minutes each)
11:00	Anne Benoit, ENS Lyon	Energy-aware mappings of series-parallel workflows onto chip multiprocessors
11:25	David  Abramson, Monash U	Assertion Based Parallel Debugging: A new way of thinking?
11:50	Aurelien  Bouteiller , UTK	Options and Challenges for Fault Tolerance
12:15	Guillaume de Verdière, CEA	Impact of Manycore on Legacy Codes
1:05  - 2:00	Lunch
2:00	Depart

Attendee List:

David	Abramson	Monash U
Pete	Beckman	ANL
Anne	Benoit	ENS Lyon
Fran	Berman	RPI
Jean-Yves	Berthou	ANR
George	Bosilca	UTK
Aurelien	Bouteiller	UTK
Franck	Cappello	Univ Paris
Guillaume	COLIN DE VERDIERE	CEA
Anthony	Danalis	UTK
Ewa	Deelman	ISI
Patrick	Demichel	HP
Luiz	DeRose	Cray
Frederic	Desprez	INRIA
Jack	Dongarra	UTK/ORNL
Geoffrey	Fox	Indiana
Al	Geist	ORNL
Andrew	Grimshaw	U Virginia
Bill	Gropp	UIUC
Torsten	Hoefler	ETH
Jeff	Hollingsworth	U Maryland
Emmanuel	Jeannot	INRIA
Thilo	Kielmann	Vrije Universiteit
Laurent	Lefevre	INRIA
Andrew	Lumsdaine	Indiana
Rusty	Lusk	Argonne
Satoshi	Matsuoka	Tokyo Institute of Technology
Doug	Miles	PGI
Bernd	Mohr	Juelich
Raymond	Namyst	U Bordeaux & INRIA
Mike	Norman	UCSD
Christian	Obrecht	INSA Lyon
Cherri	Pancake	Oregon State
Phil	Papadopoulos	UC San Diego
Jelena	Pjesivac-grbovic	Google
Padma	Raghavan	Penn State
Jonathan	Rouzaud	Lyon
Joel	Saltz	Emory U
Vaidy	Sunderam	Emory U
Frederic	Suter	CNRS/IN2P3
Martin	Swany	Indiana U
Rajeev	Thakur	Argonne
Bernard	Tourancheau	University Grenoble
Anne	Trefethen	Oxford
Jeffery	Vetter	ORNL & GATech
Xavier	Vigouroux	Bull
Malisha	Vucinic	LIG, Grenoble U
David	Walker	Cardiff U

Arrival / Departure Information:

Here is some information on the meeting in Lyon.  We have updated the workshop webpage http://bit.ly/ccdsc2012 with the workshop agenda.

On Tuesday September 11th there will be a bus to pick up participants at Lyon's Saint Exupéry (old name Satolas) Airport at 3:00. (Note that the Saint Exupéry airport has its own train station with direct TGV connections to Paris via Charles de Gaulle. If you arrive by train at Saint Exupéry airport please go to the airport meeting point (point-rencontre) (second floor, next to the shuttles, near the hallway between the two terminals, see http://www.lyon.aeroport.fr/en/maps/halls.htm#).

The bus will then travel to pick up people at the Lyon Part Dieu railway station at 4:45. (There are two train stations in Lyon, you want Part Dieu station not the Perrache station.) There will be someone with a sign at the "Meeting Point/point de rencontre" of the station to direct you to the bus.

The bus is expected to arrive at the La Maison des Contes around 5:30. We would like to hold the first session on Tuesday evening from 6:30 pm to 8:00 pm, with dinner following the session. The La Maison des Contes is about 43 Km from Lyon. For a map to the La Maison des Contes go to http://maps.google.com and type in: “427 chemin de Chanzé 69490 Dareizé” or see: Maps: http://www.cs.utk.edu/~dongarra/CCGSC-2012/maps.pdf

Map of Chateau: http://www.cs.utk.edu/~dongarra/CCGSC-2012/chateau.pdf

VERY IMPORTANT: Please send your arrival and departure times to Jack so we can arrange the appropriate size bus for transportation.  VERY VERY IMPORTANT: If your flight is such that you will miss the bus on Tuesday, September 11th at 3:00 send Bernard your flight arrival information so he can arrange for a transportation to pick you up at the train station or the airport in Lyon. It turns out that a taxi from Lyon to the Chateau can cost as much as 100 Euro and the Chateau may be hard to find at night if you rent a car and are not a French driver :-). 

At the end of the meeting on Friday afternoon, we will arrange for a bus to transport people to the train station and airport. If you are catching an early flight in the morning of Saturday, September 15th you may want to stay at the hotel located at Lyon's Saint Exupéry Airport,

see http://www.lyon.aeroport.fr/eng/en/services/index.htm for details.

There are also many hotels in Lyon area, see: http://www.lyon-france.com/

Due to room constraints at the La Maison des Contes, you may have to share a room with another participant. Dress at the workshop is informal.  Please tell us if you need special requirements (vegetarian food etc...) We are expecting to have internet and wireless connections at the meeting.

Please send this information to Jack (dongarra@eecs.utk.edu) by July18^th.

Name:

Institute:

Title:

Abstract:

Participant’s brief biography:

Arrival Details:

1.	David	Abramson	Monash U	VEGATERIAN meet at train station 2:56 from Paris; depart Friday PM
2.	Pete	Beckman	ANL	Arrive 1:35 LH (UN8916)
3.	Anne	Benoit	ENS Lyon	drive
4.	Fran	Berman	RPI	Arrive train station not airport (Cherri) airport KLM9288 11:15; depart Saturday 7:35am AF8611
5.	Jean-Yves	Berthou	ANR	Drive
6.	George	Bosilca	UTK	Take bus from airport on AF78-5 at 2:10
7.	Aurelien	Bouteiller	UTK	Take bus from Part-Dieu
8.	Franck	Cappello	Univ Paris	Take bus from Part Dieu Return on Friday to Part Dieu
9.	Guillaume	Colin de Verdiere	CEA	Take bus from Part Dieu Return on Friday to Part Dieu
10.	Anthony	Danalis	UTK	Arrive airport 2:30 AF7644
11.	Ewa	Deelman	ISI	Drive (rent a car) Arrive late
12.	Patrick	Demichel	HP	Take bus from Part Dieu
13.	Luiz	DeRose	Cray	Drive
14.	Frederic	Desprez	INRIA	Drive
15.	Jack	Dongarra	UTK/ORNL	Arrive early; depart Saturday from airport
16.	Rafael	Ferreira da Silva	INSA-Lyon	Drive
17.	Geoffrey	Fox	Indiana	Arrive airport 2:30 DL8344; depart Saturday 7:35am DL8611
18.	Al	Geist	ORNL	Arriving airport 11:35 DL8414; departing Saturday DL8611
19.	Andrew	Grimshaw	U Virginia	Arriving LH 2248 12:10, depart Friday evening 5:25
20.	Bill	Gropp	UIUC	Arrive at airport AF7644 2:30; depart Saturday 7:35am flight
21.	Torsten	Hoefler	ETH	Arrive 3:22pm at Lyon-Part-Dieu; depart Friday Part-Dieu
22.	Jeff	Hollingsworth	U Maryland	arrive at airport; leave early Friday morning
23.	Emmanuel	Jeannot	INRIA	Late: Wednesday Arrive airport 7:55 am from Bordeaux; depart 18:30.
24.	Thilo	Kielmann	Vrije Universiteit	Traveling by train. Arriving on Tuesday, at 4pm at Lyon Part Dieu station.
25.	Laurent	Lefevre	INRIA	VEGETARIAN Driving; leaving Thursday morning
26.	Andrew	Lumsdaine	Indiana	VEGETARIAN Bus from airport; departure Saturday
27.	Rusty	Lusk	Argonne	Arrive 1:35 LH (UN8916)
28.	Satoshi	Matsuoka	TITech	Late: Wednesday Sep. 12, 9:50 AM FRA-LYS (LH1074) Depart Saturday
29.	Doug	Miles	PGI	Drive
30.	Bernd	Mohr	Juelich	Meet bus at train station
31.	Raymond	Namyst	U Bordeaux & INRIA	Meet bus at airport and back on Friday to airport
32.	Mike	Norman	UCSD	Arrive 1:35 AA8916 depart on Saturday UA8891
33.	Christian	Obrecht	INSA Lyon	Drive
34.	Cherri	Pancake	Oregon State	Meet bus at Part Dieu on Tuesday;
35.	Phil	Papadopoulos	UC San Diego	Airport Arrive 2:30 DL 8444 depart Friday morning 10:50 am
36.	Jelena	Pjesivac-Grbovic	Google	Arrive Tues @ 1:35 on LH1076 Depart Sunday
37.	Padma	Raghavan	Penn State	Airport pickup and return
38.	Jonathan	Rouzaud	Lyon	Part Dieu?
39.	Joel	Saltz	Emory U	Airport pickup and return
40.	Vaidy	Sunderam	Emory U	Join bus at the airport and return to airport on Friday.
41.	Frederic	Suter	CNRS/IN2P3	VEGATERIAN Drive
42.	Martin	Swany	Indiana U	Airport pickup and return
43.	Rajeev	Thakur	Argonne	Arrive by train Part Lieu pm depart by train on Friday at 5pm
44.	Bernard	Tourancheau	University Grenoble	Bus
45.	Anne	Trefethen	Oxford	Arrive at airport 2:30 AF7644; depart Friday from Part-Dieu at 3:00
46.	Jeffery	Vetter	ORNL & GATech	Tuesday via Paris DL8414 11:35am; depart Lyon Saturday 7:30 DL 8611
47.	Xavier	Vigouroux	Bull	Drive; leave Thursday
48.	Malisha	Vucinic	LIG Grenoble	Drive
49.	David	Walker	Cardiff U	airport; arriving 11:15am

Abstracts:

David Abramson, Monash University

Title:

Assertion Based Parallel Debugging: A new way of thinking?

Abstract:

Programming languages have advanced tremendously over the years, but program debuggers have hardly changed. Sequential debuggers do little more than allow a user to control the flow of a program and examine its state. Parallel ones support the same operations on multiple processes, and are adequate with a small number of cores, but become unwieldy and ineffective on very large machines. Typical scientific codes have enormous multi-dimensional data structures and it is impractical to expect a user to view the data using traditional display techniques.

In this talk I will discuss the use of debug-time assertions (both within and across programs), and show that these can be used to debug parallel programs. The techniques reduce the debugging complexity because they reason about the state of large arrays without requiring the user to know the expected value of every element. When used across programs, the technique can help find errors that occur when a program is ported to a new platform. Whilst assertions can be expensive to evaluate, their performance can be improved by running them in parallel. We have implemented these ideas in a new debugger called Guard, and will illustrate its performance  on tens of thousands of cores on a Cray XE6.

Anne BENOIT, ENS Lyon and Institut Universitaire de France

Title:

Energy-aware mappings of series-parallel workflows onto chip multiprocessors

Abstract:

We discuss the problem of mapping streaming applications that can be modeled by a series-parallel graph onto a 2-dimensional tiled CMP architecture. The objective of the mapping is to minimize the energy consumption, using dynamic voltage scaling techniques, while maintaining a given level of performance, reflected by the rate of processing the data streams. This mapping problem turns out to be NP-hard, but we identify a simpler instance, whose optimal solution can be computed by a dynamic programming algorithm in polynomial time. Several heuristics are proposed to tackle the general problem, building upon the theoretical results. Finally, we assess the performance of the heuristics through a set of comprehensive simulations. 

Fran Berman, RPI

Title: The Digital Society

Abstract:

In the 18^th and 19^th centuries, new technologies fueled an Industrial Revolution that dramatically changed the way people lived and worked. In the 20^th and 21^stcenturies, digital information and its infrastructure has fueled an Information Age, creating new capabilities and new approaches to how we operate, communicate, and interrelate.

Achieving the potential of the Information Age requires more than technical solutions. Critical social and organizational structures facilitating digital economics, legal and policy frameworks, social behavior and “lawfulness” in virtual environments, and overarching governance and community interaction are required to take full advantage of the digital bounty now available. In this talk, we touch on the challenges and opportunities in creating a digitally-fueled society that meets its tremendous potential for success and impact.

George  Bosilca, UTK

Title: PaRSEC: A Modular Framework for Performance and Productivity on Hybrid Architectures

Abstract:

The convergence of several unprecedented challenges -- formidable new design constraints, revolutionary amounts of heterogeneity, and a complete transformation of the programming paradigm -- suggest that traditional SPMD message passing programming model became unsuitable for emerging hardware designs and that the “application-architecture performance gap,” will become more difficult than ever to close.

A dataflow model has the capability to tolerate memory and synchronization latencies while exposing the maximum degree of parallelism from algorithms, thanks to an explicit depiction of data dependencies. In this talk I will introduce the PaRSEC runtime, a low level data centric execution environment, and one of it's domain specific programming paradigm. The presented results show that this approach has the potential to unleash an extreme level of performance when compared to legacy programming approaches, and while exhibiting a natural porting of the same algorithm to a variety of hardware platforms, encompassing heterogeneous resources such as GPU accelerators, with a steady and predictable delivery of performance.

Aurelien  Bouteiller, UTK

Title:

Abstract:

Franck Cappello, Univ Paris

Title: Failure prediction: current situation and open questions

Abstract:

One form of failure avoidance relies on runtimes, OS, and system management environments to perform failure prediction and proactive actions. Making failure avoidance from failure prediction a credible solution for Exascale resilience supposes (i) to improve failure prediction drastically and (ii) to adapt the relevant software layers for failure avoidance. This talk will focus on the most critica

l part: failure prediction. Failure prediction quality is defined by (i) precision, (ii) coverage (recall) and (iii) enough time lags between correctly predicted failures and the real failures to trigger and perform proactive actions like migration, replication or checkpointing.  In the past few years, several key results have demonstrated that new anomaly/symptom detection and correlation analysis algorithms can provide precise on-line failure prediction in HPC systems. The time lag observed for the most efficient prediction approaches is consistent with the time taken by fast proactive actions, like checkpointing on local SSD or migration. However the proportion of failures that could be predicted over all observed failures is still low and stays below 50% even for the most advanced prediction approaches. The objective is then to improve the failure prediction coverage from 50% toward 80% or 90%, in HPC systems. We will present the latest approaches in failure prediction and discuss what we think are the sources of low coverage.

Guillaume Colin de Verdiere, CEA

Title: Impact of manycore on legacy codes

Abstract:

In this talk we'll emphasize why the community is facing a major rewrite of the legacy codes to cope with the new forthcoming architectures.

Anthony Danalis, UTK

Title: Task data flow analysis and extraction from serial input code

Abstract:

The DAGuE system provides an efficient platform for efficient distributed execution of task-based applications at scale.  However, developing applications directly in the task data flow language of DAGuE can be challenging.  In this talk I will outline the compiler technology we have developed to analyze the serial, loop-based codes written for the PLASMA project and convert them into DAGuE's internal representation of the task data flow.  I will outline the analysis steps, discuss theoretical results and limitations, and present ideas for future directions.

Ewa Deelman, USC Information Sciences Institute

Title: Hosted Science: How to support complex scientific applications on the cloud

Abstract:

Scientists today are exploring the use of new tools and computing platforms to do their science.  They are using workflow management tools to describe and manage complex applications and are evaluating the features and performance of clouds to see if they meet their computational needs.  Although today, hosting is limited to providing virtual resources and simple services, one can imagine that in the future entire scientific analysis will be hosted for the user. The latter would specify the desired analysis, the timeframe of the computation, and the available budget. Hosted services would then deliver the desired results within the provided constraints. 

Frequently, applications are not represented by single workflows but rather as sets of related workflows—workflow ensembles. Thus, hosted services need to be able to manage entire workflow ensembles, evaluating tradeoffs between completing as many high-value ensemble members as possible and delivering results within a certain time and budget. The talk presents a range of algorithms that provision and schedule workflow ensembles while satisfying the user-defined budget and deadline constraints.  The algorithms are evaluated via simulation using a set of realistic scientific workflow ensembles.

Patrick Demichel, HP

Title: HP-labs researches for Exascale

Abstract:

It is expected that Exascale will present many challenges to our industry. The HP-labs are conducting extensive and holistic researches to respond to the challenges to enable the design of a platform that will be economically viable, very efficient in term of power and easy to manage and program.This talk will expose where we are in our studies, why we do those efforts  and what benefits and impacts we expect from those some disruptive technologies that are now moving in the development phase.

Luiz DeRose, Cray

Title: Productive Performance on Accelerator Based Systems Using OpenACC Compilers and Tools

Abstract:

One of the critical hurdles for the widespread adoption of accelerated computing in high performance computing is programming difficulty. To be an effective HPC platform, accelerator based systems need a high level software development environment to facilitate the porting and development of applications, so they can be portable and run efficiently on either accelerators or CPUs. In this talk I will present Cray's high level parallel programming environment for accelerator based systems, which consists of tightly coupled compilers, libraries, and tools that can interoperate and hide the complexity of the system. Ease of use is possible with compiler making it feasible for users to write applications in Fortran, C, or C++ with OpenACC directives, tools to help users port, debug, and optimize for accelerators, as well as conventional multi-core CPUs; and auto-tuned scientific libraries.

Frederic Desprez, INRIA

Title:

Abstract:

Geoffrey Fox, Indiana

Title: Robust Parallel Clustering Algorithms

Abstract:

We describe a family of robust clustering algorithms and lessons from their use and parallelization on both HPC and Cloud Platforms. We present several applications and comparisons with other clustering approaches. We discuss the situation where points (to be clustered) lie in metric and non metric spaces and how in latter case one can develop O(NlogN) algorithms rather than naive O(N2) algorithms. We note that there is always a straightforward parallelism over points but that in some cases parallelism over clusters must be exploited for efficiency and it leads to a very different communication structure. We describe work to be done to create a full family of scalable high performance clustering approaches.

Al Geist, ORNL

Title: Distributed Peer-to-Peer Control in the Cloud

Abstract: Resilience has become a critical need for multi-petaflop Leadership systems and large Clouds. This talk will present a symmetric peer-to-peer distributed control algorithm that can be used in large parallel runtime environments. We will describe how it can do fast adaptation of resources and maintain system-wide state despite multiple points of failure at high fault rates. These will be important features for future Leadership and Cloud systems.

Andrew Grimshaw, U Virginia

Title: The Global Federated File System: Early Results

Abstract:

Federated, secure, standardized, scalable, and transparent mechanism to access and share resources, particularly data resources, across organizational boundaries that does not require application modification and does not disrupt existing data access patterns has been needed for some time in the computational science community. The Global Federated File System (GFFS) addresses this need and is a foundational component of the NSF-funded eXtreme Science and Engineering Discovery Environment (XSEDE) program. The GFFS allows user applications running on campus and research group machines to directly access (create, read, update, delete) files and other resources at NSF-funded service provider sites and collaborator sites as if they were in their local operating system file system. Similarly, the GFFS allows applications running at the service providers to directly access files on campus or research group resources as if they were located at the center.  Existing applications, whether they are statically linked binaries, dynamically linked binaries, or scripts (shell, PERL, Python), can access resources anywhere in the GFFS without modification (subject to access control). This paper describes the GFFS, its goals, use cases, and limitations.

Torsten Hoefler, ETH Zurich

Title: Topology Mapping for Petascale++ Supercomputers

Abstract: The steadily increasing number of nodes in high-performance computing systems and the technology- and power-constraints in networking lead to sparse large-scale networks. Efficient mapping of application communication patterns to such sparse topologies gains importance as systems grow to petascale and beyond. Such topology mappings are supported in parallel programming frameworks such as MPI, but are often not well implemented. We show that the topology mapping problem is NP-complete and analyze and compare different practical topology mapping heuristics. We demonstrate an efficient and fast new heuristic which is based on graph similarity and show its utility with application communication patterns on real topologies. Our mapping strategies support heterogeneous networks and show significant reduction of congestion on torus, fat-tree, and the PERCS/Dragonfly network topologies for irregular problems. We also demonstrate that the benefits of topology mapping grow with the network size and show how our algorithms can be used in a practical setting to optimize communication performance. We argue that maximum congestion and average dilation are good metrics for application performance and network power consumption, respectively. Our efficient topology mapping strategies are shown to reduce network congestion by up to 80%, reduce average dilation by up to 50%, and improve benchmarked communication performance by 18%.

Jeff Hollingsworth, U Maryland

Title: Automatically Adapting Programs for Mixed-Precision Floating-Point Computation

Abstract:

As scientific computation continues to scale, it is crucial to use floating-point arithmetic processors as efficiently as possible. Lower precision allows streaming architectures to perform more operations per second and can reduce memory bandwidth pressure on all architectures.  However, using a precision that is too low for a given algorithm and data set will result in inaccurate results.

In this talk I will present a framework that uses binary instrumentation and modification to build mixed-precision configurations of existing programs that were originally developed to use only double precision.  This allows developers to easily experiment with mixed-precision configurations without modifying their source code, and it permits auto-tuning of floating-point precision. I will also present some results from a related tool that helps identify when floating point cancellation has occurred.

Emmanuel Jeannot, INRIA

Title: Process placement of MPI an multi-threaded applications. Why, how and for what gain?

Abstract:

Standard programming models such as MPI or pthread make almost no assumption on the way the machine running the application is organized.  However, machine topology may have a great impact on the application performance. In this talk we will discuss several examples on how process and thread placement can be done and the gain that can be achieved thanks to a careful placement.

Abstract:

Thilo Kielmann, Vrije Universiteit, Amsterdam

Title:

Task Farming in Cloud Environments under Budget Constraints

Abstract:

Elastic applications like bags of tasks can greatly benefit from Infrastructure as a Service (IaaS) clouds that let users allocate compute resources on demand, charging based on reserved time intervals. Users, however, still need guidance for mapping their applications onto multiple IaaS offerings, both minimizing execution time and respecting budget limitations.  For budget-controlled execution of bags of tasks, we have developed our BaTS scheduler. BaTS estimates possible budget and makespan combinations using a tiny task sample, and then executes a bag within the user’s budget constraints. Previous work has shown the efficacy of this approach.

In this presentation, I will present recent developments, among which are the use of paid-for-but-idle machines during the tail phase of an execution and the use of spot market instances.

Laurent Lefever, INRIA

Title:

Energy efficiency in HPC and exascale

Abstract :

As energy prices rise and environmental issues begin to be taken into account in the HPC world; this talk will review some latest advances in energy efficiency concerning HPC infrastructures and services.

Andrew Lumsdaine, Indiana

Title: Scalable Graph Algorithms Based on Active Messages

Abstract:

Abstract: Data-driven applications, such as graph analytics, are unique in that the computational structure of the applications is determined by the input data.  Static analysis is difficult as a result, necessitating dynamic, lightweight, execution-time solutions. Version 2.0 of the Parallel Boost Graph Library (PBGL) demonstrates one approach to building a modular, scalable, and perhaps most importantly, performance portable set of graph kernels. The PBGL 2.0 uses the AM++ active messaging library (an implementation of the Active Pebbles programming and execution model) to provide portable, generic, thread-safe messaging support on a variety of platforms. On top of AM++, PBGL 2.0 provides a variety of graph types, auxiliary data structures, and algorithmic kernels suitable for both shared- and distributed-memory parallelism (e.g. threads and processes) with the potential for straightforward extension to various types of accelerators.

Doug Miles, PGI

Title: The Challenges of Portable Heterogeneous Programming

Abstract:

Most CPU-only large-scale systems have a similar look-and-feel: many homogeneous nodes communicating via MPI, each node has a few identical processor chips, each chip has multiple identical cores, each core has some SIMD processing capability.  Programming one such system is very like programming any other, regardless of chip vendor, number of cores, SIMD width, interconnect fabric, etc.  However, the coming onslaught of heterogeneous node types presents interesting challenges to application writers and programming tools vendors.  Here, we start by exploring the growing space of heterogeneous architectures, showing key similarities and differences.  We summarize many of the problems that will arise for programmers when developing and tuning applications, specifically when striving for performance portability, and argue that productivity requires an appropriate programming environment.  We close with the strategies that PGI is pursuing to address these challenges.

Raymond Namyst, U Bordeaux & INRIA

Title: Programming heterogeneous, accelerator-based multicore machines: a runtime system's perspective

Abstract:

Heterogeneous accelerator-based parallel machines, featuring manycore CPUs and with GPU accelerators provide an unprecedented amount of processing power per node. Dealing with such a large number of heterogeneous processing units -- providing a highly unbalanced computing power -- is one of the biggest challenge that developpers of HPC applications have to face. To Fully tap into the potential of these heterogeneous machines, pure offloading approaches, that consist in running an application on regular processors while offloading part of the code on accelerators, are not sufficient.

In this talk, I will go through the major software techniques that were specifically designed to harness heterogeneous architectures, focusing on runtime systems. I will discuss some of the most critical issues programmers have to consider to achieve portability of performance, and I will show how advanced runtime techniques can speed up applications in the domain of dense linear algebra.

Eventually, I will give some insights about the main challenges designers of programming environments will have to face in upcoming years.

Mike Norman, UCSD

Title: Early experiences with Gordon – a flash-based cluster for data-intensive applications

Abstract:

In Feb 2012 SDSC Gordon went into production for US academic researchers as an XSEDE resource. Gordon is a 300 TF, 1024 node dual Sandy Bridge node compute cluster with several non-standard features targeting data-intensive applications: 300 TB of flash SSD for random IO-intensive applications, and ScaleMP’s virtual shared memory software, for memory-intensive applications. In this talk I report on early user experiences with Gordon, as well as some of the operational and reliability experiences we have encountered deploying these new technologies. Gordon is supported by the National Science Foundation under Cooperative Agreement OCI-0910847.

Christian Obrecht, Institute: EDF R&D / CETHIL UMR5008

Title: Efficient GPU cluster implementation of the lattice Boltzmann method

Abstract:

The lattice Boltzmann method (LBM) is an innovative and promising approach in computational fluid dynamics. From an algorithmic standpoint it reduces to a regular data parallel procedure and is therefore, well-suited to high performance computations. Numerous works report efficient implementations of the LBM for the GPU, but very few mention multi-GPU versions and even fewer GPU cluster implementations. Yet, to be of practical interest, GPU LBM solvers need to be able to perform large scale simulations. In the present contribution, we describe an efficient LBM implementation for CUDA GPU clusters. Our solver consists of a set of MPI communication routines and a CUDA kernel specifically designed to handle three-dimensional partitioning of the computation domain.  Performance measurements were carried out on a small cluster. We show that the results are satisfying, both in terms of data throughput and parallelisation efficiency.

Cherri Pancake, Oregon State

Title: Cloud-Enabled Collaboration:  Can We Learn from Past Mistakes?

Abstract:

The build-out of cloud-based computing and data storage will make it possible to support research collaboration of unprecedented types and scales.  Shifting the emphasis from physical resources to services helps researchers re-think how they approach problems.  Instead of focusing on "what my group can accomplish using our facilities," potential cloud users can envision what might be achieved with open access to global data and computing power.  The amorphous, unstructured nature of the cloud, however, raises questions about what level of usability and usefulness will actually be achieved. ?Obvious? resource-sharing strategies have been tried in the last two decades, but most have fallen short of user expectations and needs. Creative new solutions for sharing models, tools, and research data are called for, informed by consideration of how users measure success. This presentation will review the lessons of recent efforts to enable/protect resources shared across distributed and interdisciplinary communities, and suggest how they could be leveraged in a cloud environment.

Phil  Papadopoulos, UC San Diego

Title: International Cloud Interoperation Experiments in PRAGMA

Abstract:

The Pacific Rim Applications and Grid Middleware Assembly is a 30+ institution grass roots or bottom’s up experimentation  and computing infrastructure. Originally envisioned as a mutli-institutional grid, this persistent environment is evolving to a multi-cloud where the executable is a virtual machine instead of an application executable.  With partners at AIST (Japan) and NCHC (Taiwan) we have developed semi-automated techniques to convert application-specific virtual machine images to run in multiple cloud hosting infrastructures, including Amazon EC2.   VM’s are uploaded to distributed, strongly authenticated, and replicated storage infrastructure based on gFarm and then copied out and translated to local hosting infrastructure.  This talk will describe the process, the challenges such as overlay networking, VM control and a path forward to creating usable infrastructure for small groups of collaborating teams. We will walk through two different use models that operate across different cloud infrastructures to illustrate various components.

Jelena Pjesivac-grbovic, Google

Title: Google Compute Engine: Google solution for large scale data processing and analytics

Abstract:

Padma Raghavan, Penn State

Title: Do Fixed-Problem Speedups Matter?

Abstract:

In parallel scientific computing, we have often focused on weak-scaling, i.e., maintaining iso-efficiency while increasing the size of the problem with the number of processors. This is especially true for sparse scientific computing on matrices, meshes and graphs that typically involve irregular data access patterns. Now, fixed-problem speedups can be quite low even for a relatively small number of processors. We characterize the speedup and system energy profiles of such sparse applications in an effort to understand and close the gaps between ideal and achievable efficiencies. We first demonstrate how speedup profiles can be used in the scheduling of workloads of sparse applications to obtain substantial decreases in time to completion and savings in system energy. Next, we demonstrate how data-access patterns  can be used in NUMA-aware  dynamic scheduling to deliver higher fixed-problem speed-ups while saving system energy compared to traditional schemes in OpenMP. We conclude with a discussion of some open problems related to scheduling to enhance fixed problem speedups.

Joel Saltz, Emory U

Title: Toward Derivation, Management, and Analysis of Exascale Feature Sets

Abstract:

Integrative analyses of large scale temporal spatial datasets play increasingly important roles in many areas of science and engineering.  We outline middleware infrastructure requirements to extract, manage and analyze broadly descriptive, reproducible and semantically meaningful information from very large image, sensor or simulation datasets. This is both a compute and data intensive problem. The need for descriptiveness motivates the need to run complementary multi-scale analysis algorithms while the need for reproducibility drives the need to run many different algorithms aimed at extracting the same information.  The analyses derive many very large semantically complex feature sets that must be repeatedly managed, queried and re-analyzed.

This talk will describe both motivating application scenarios involving complementary digital microscopy,  Radiology and “omic” analyses in cancer research and will describe the systems software effort being launched to address the challenges posed by this kind of integrative analysis.

Vaidy Sunderam, Emory University

Title: Towards Computing as a Utility via Adaptive Middleware

Abstract:

With the increasing maturation of grids and clouds, the vision of computing as a utility, is starting to become a reality. However, usability of such platforms for executing science and engineering applications is a challenge. In order for these applications to utilize resources ranging from on-premise to federated to on-demand platforms, they often need target-specific adjustments and reconciliations -- which pose considerable logistical and efficiency obstacles.  This talk will describe the proposed design of adaptive middleware to enhance the executability of applications on varied computational back-ends. Preliminary experiences with use cases that measure the feasibility and issues with cross platform computing will be discussed.

Fredric Suter,  CNRS/IN2P3

Paving the Road for the Simulation of Exascale and Cloud Systems.

Abstract:

The last decade has brought tremendous changes to the characteristics of large scale distributed computing platforms. Large grids processing terabytes of information a day and the peer-to-peer technology have become common. Infrastructures of the next decade, such as petaflop/exaflop computers and emerging clouds are even more challenging. Understanding how to efficiently use such platforms is difficult and simulation has proved to be a very effective approach for their study. In this talk, I will detail the main challenges that a simulation tool has to face when it targets large scale distributed platforms and which technical and methodological solutions were proposed in the SimGrid simulation framework for Grid and P2P systems. I will also present how this framework will be extended to address Clouds and peta/exascale systems.

Martin Swany, Indiana U

Title: High Performance Data Transfer with RDMA in the Wide Area

Abstract:

The effective and efficient utilization of wide-area networks is a persistent challenge in today's network ecosystem. The trend towards increasingly distributed computing environments supporting data-intensive applications only compounds the issue, making data movement a cross-cutting concern, affecting users of distributed computing and data-intensive science alike.  One promising development is the use of zero-copy RDMA.  This talk discusses an evaluation of RDMA over Ethernet.

Rajeev Thakur, Argonne

Title: Challenges for Communication Libraries and Runtime Systems at Exascale

Abstract:

Communication libraries and runtime systems for exascale will have to deal with the challenges of hundreds or thousands of cores per node, limited memory per core, heterogeneous cores and memory, millions of processes/threads in total, frequent failures, and complex hardware topologies. This talk will present current work being done at Argonne to address some of these challenges

Anne Trefethen, Oxford

Title: Enabling Science using the SKA

Abstract: The Square Kilometre Array (SKA) is the next generation radio telescope. It will allow a sensitivity of more than 50 times that of existing telescopes, and 10,000 times the survey speed and will provide data to answer fundamental science questions on gravitation and magnetism, galaxy formation and even the question of life on other planets The SKA provides a fabulous information technology challenge with a typical data rate from each dish antenna on the order of 100Gbs-1 aggregating to over 100Tbs-1  and a need for Exaflop computation  for post-processing.

The infrastructure required to support the various science cases will need to range from real-time capability to transport and analyse data at high-data rates and the capacity to store and “publish” the data for later analysis and interpretation by the global astrophysics community.  The computational systems will likely range from specifically designed FPGA-like units to exascale computing and Cloud-like data centres with energy as a major constraint on the design though out.

In this presentation I will talk about some of the design simulations and the pipelines for real-time analysis that have been built for pathway systems and that have already resulted in new science results.

Jeffery Vetter, ORNL & GA Tech

Title:  Aspen: a performance modeling language for codesigning Exascale applications and architectures

Abstract:  We present a new approach to analytical performance modeling using Aspen, a domain specific language. Aspen (Accurate Scalable Performance Engineering Notation) fills an important gap in existing performance modeling techniques and is designed to enable rapid exploration of new algorithms and architectures.  It includes a formal specification of an application’s performance behavior and an abstract machine model. We provide an overview of Aspen’s features and demonstrate how it can be used to express a performance model for a three dimensional Fast Fourier Transform. We then demonstrate the composability of Aspen in other models. For example, we import and reuse this FFT model into a molecular dynamics model. Given the systematic approach provided by Aspen, we have also created a number of tools that allow scientists to balance application and architecture factors quickly and systematically.

David W. Walker, Cardiff University, UK

Title: Performance at the Exascale: a Simple Model for Heterogeneous Platforms

Abstract: The next decade will see substantial and rapid changes in the architecture of high-end computing platforms, with future supercomputers likely to consist of multiple nodes, each of which will be a heterogeneous mix of multicore processors and attached GPUs. This trend is expected to lead to an exascale computer with an energy budget of about 20 MW by the early 2020’s. To achieve this goal requires innovations in areas such as energy/power management, fault tolerance and reliability, and programmability. This talk examines how a simple performance model of heterogeneous high-end computing platforms can inform the design of such machines, and help address the challenges faced by the computational science community in seeking to harness of the order of millions of cores, running billions of threads, to solve scientific applications. This simple performance model is useful in assigning computational work to the different processing devices in a heterogeneous HPC system, and also illustrates the distinction between compute-bound and memory-bound machine/algorithm combinations. The performance of some algorithms that are widely used in scientific applications will be examined in order to test the feasibility of achieving exascale performance in practice.

Biographies of Attendees

David  Abramson

Institute: Monash U

Professor David Abramson has been involved in computer architecture and high performance computing research since 1979. Previous to joining Monash University in 1997, he has held appointments at Griffith University, CSIRO, and RMIT. At CSIRO he was the program leader of the Division of Information Technology High Performance Computing Program, and was also an adjunct Associate Professor at RMIT in Melbourne. He served as a program manager and chief investigator in the Co-operative Research Centre for Intelligent Decisions Systems and the Co-operative Research Centre for Enterprise Distributed Systems.

Abramson is currently Director of the Monash e-Education Centre , science director of the Monash e-Research Centre and a Professor of Computer Science in the Faculty of Information Technology at Monash University, Australia. He is a fellow of the Association for Computing Machinery (ACM) and the Academy of Science and Technological Engineering (ATSE), and a Senior Member of the IEEE.

Abramson has served on committees for many  conferences and workshops, and has published over 200 papers and technical documents. He has given seminars and received awards around Australia and internationally and has received over $8 million in research funding.

He also has a keen interest in R&D commercialization and consults for Axceleon Inc, who produce an industry strength version of Nimrod, and Guardsoft, a company focused on commercialising the Guard relative debugger.

Abramson’s current interests are in high performance computer systems design and software engineering tools for programming parallel, distributed supercomputers  and stained glass windows.

Pete Beckman

Institute: ANL

Anne BENOIT

Institute: ENS Lyon and Institut Universitaire de France

Anne Benoit received the PhD degree from Polytechnical Institute of

Grenoble (INPG) in 2003, and the HDR from Ecole Normale Superieure(ENS Lyon) in 2009. She is currently an associate professor in the Computer Science Laboratory LIP at ENS Lyon, France. She is the author of 24 papers published in international journals, and 56 papers published in international conferences. She is the advisor of 4 PhD theses. Her research interests include algorithms design and scheduling techniques for parallel and distributed platforms, and also the performance evaluation of parallel systems and applications. She is Associate Editor of JPDC. She was the program chair of several workshops and conferences. She is a senior member of the IEEE.  She has been elected a Junior Member of Institut Universitaire de France in 2009.

Fran Berman

Institute: RPI

Dr. Francine Berman is Professor of Computer Science at Rensselaer Polytechnic Institute. She is a Fellow of the Association of Computing Machinery (ACM) and a Fellow of the IEEE. In 2009, Dr. Berman was the inaugural recipient of the ACM/IEEE-CS Ken Kennedy Award for “influential leadership in the design, development, and deployment of national-scale cyberinfrastructure.” 

Prior to joining Rensselaer, Dr. Berman was Professor in the UC San Diego Department of Computer Science and Engineering, and first holder of the High Performance Computing Endowed Chair in the Jacobs School of Engineering. From 2001 to 2009, Dr. Berman served as Director of the San Diego Supercomputer Center (SDSC) where she led a staff of 250+ interdisciplinary scientists, engineers, and technologists. From 2009 to 2012, she served as Vice President for Research at Rensselaer Polytechnic Institute, stepping down from administrative duties in 2012 to take an expanded role within the data community. 


Dr. Berman has served on a broad spectrum of national and international leadership groups and committees including the National Science Foundation's Engineering Advisory Committee, the National Institutes of Health's NIGMS Advisory Committee, the Anita Borg Institute for Women and Technology Board of Trustees, the National Academy of Sciences Board on Research Data and Information, the U.S. President's Council of Advisors on Science and Technology (PCAST) Networking and Information Technology Research and Development (NITRD) Review Working Group, and others. From 2007-2010, she served as co-Chair of the US-UK Blue Ribbon Task Force for Sustainable Digital Preservation and Access. Dr. Berman is currently Chair of the Information, Computing and Communication Section (Section T) of the American Association for the Advancement of Science (AAAS), vice-Chair of the Anita Borg Institute Board of Trustees, and co-Chair of the National Academies Board on Research Data and Information (BRDI).

Jean-Yves Berthou

Institute: ANR

Jean-Yves Berthou has joined in September 2011 the French National Research Agency (ANR) as Director of the Department for Information and Communication Science and Technologies. He has been before that the Director of the EDF R&D Information Technologies program since 2008 and the coordinator of the EESI European Support Action, European  Exascale Software Initiative, www.eesi-project.eu.

Jean-Yves received a Ph.D in computer science from "Pierre et Marie Curie" University (PARIS VI) in 1993. His research deals mainly with Parallelization, Parallel Programming and Software Architecture for Scientific Computing.

Jean-Yves joined EDF R&D in 1997 as a researcher. He was the head of the Applied Scientific Computing Group (High Performance Computing, Simulation Platforms Development, Scientific Software Architecture) at EDF R&D from 2002 to 2006. He has been Chargé de Mission – Strategic Steering Manager for Simulation, in charge of the simulation program at EDF R&D from 2006 to 2009.

George  Bosilca

Institute: UTK

George Bosilca is a Research Associate Professor of Electrical Engineering and Computer Science, University of Tennessee, Knoxville. He obtained his PhD from the University of Paris XI with a background in computer architecture and parallel computing. Active member of several large scale projects (Open MPI, MPICH-V, CCI, DAGuE), his research encompass a large area in the distributed computing world. From low level network protocols to algorithmic based fault tolerance, Dr. Bosilca research targets to reduce the gap between peak and sustained performance on large scale execution environments, maximizing the performance of the heterogeneous current and future computing platforms.

Aurelien  Bouteiller

Institute: UTK

Aurelien Bouteiller received is Ph.D. from University of Paris in 2006, under the direction of Franck Cappello. His research is focused on improving performance and reliability of distributed memory systems. Toward that goal, he investigated automatic (message logging based) checkpointing approaches in MPI, Algorithm Based fault tolerance approaches and their runtime support, mechanisms to improve communication speed and balance within nodes of many-core clusters, and employing emerging data flow programming models to negate the raise of jitter on large scale systems (DAGuE project). These works resulted in over twenty-five publications in international conferences and journals and three distinguished paper awards from IPDPS and EuroPar. He his also a contributor to Open MPI and participates to the MPI-3 Forum.

Franck Cappello

Institute: Univ Paris

Franck Cappello holds a research director position at INRIA and is visiting professor in Computer Science at University of Illinois at Urbana Champaign. Since 2009, he is co-director with Marc Snir of the INRIA-Illinois Joint-Laboratory on PetaScale Computing (http://jointlab.ncsa.illinois.edu/) where he is also leading the Resilience/Fault Tolerance effort. He is leading the roadmaping effort on Resilience/Fault Tolerance for EESI2 (European Exascale Software Initiative) and led similar effort for IESP (International Exascale Software Project : http://www.exascale.org) and EESI1.  He is the main PI of the G8 ECS (Enabling Climate Simulation at Exascale: https://wiki.engr.illinois.edu/display/G8/) project gathering researchers from USA, France, Germany, Japan, Canada and Spain with the objective of identifying scalability, performance and resilience solutions for running CESM at extreme scale.

Anthony Danalis

Institute: UTK

Anthony Danalis is a Research Associate at the Innovative Computing Laboratory at the University of Tennessee. He did his undergraduate studies in Physics at the University of Crete, and his graduate studies in Computer Science at the University of Crete, and the University of Delaware.  His PhD research focused on Compiler optimizations for High Performance Computing and this has been his main area of interest ever since.

Guillaume de Verdière

Institute: CEA

Senior expert at CEA, Guillaume Colin de Verdiere is investigating new architectures (GPU, MIC, ...)  which might be good candidates as building blocks for CEA's next generations of supercomputers.

Ewa Deelman

Institute: USC Information Sciences Institute

Patrick Demichel

Institute: HP

1975 : computer architecture degree from Control Data Corporation Institute Paris Since : work for Hewlett Packard : 32 years of working on scientific business : HP1000 Real time computing; then HP9000 hard+soft+linux, then 5 years on development of IA64 "US labs", since 8 years "senior HPC architect"  with focus on largest projects in EMEA  Always a focus on hard/soft interactions and performance tuning; from CPU to IO disk/IB  etc Now  a big focus on emerging technologies : sensors, memristors, photonics, low power etc.

Luiz DeRose

Institute: Cray

Frederic Desprez

Institute: INRIA

Jack Dongarra

Institute: UTK/ORNL

Jack Dongarra holds an appointment at the University of Tennessee, Oak Ridge National Laboratory, and the University of Manchester. He specializes in numerical algorithms in linear algebra, parallel computing, use of advanced-computer architectures, programming methodology, and tools for parallel computers. He was awarded the IEEE Sid Fernbach Award in 2004 for his contributions in the application of high performance computers using innovative approaches; in 2008 he was the recipient of the first IEEE Medal of Excellence in Scalable Computing; in 2010 he was the first recipient of the SIAM Special Interest Group on Supercomputing's award for Career Achievement; and in 2011 he was the recipient of the IEEE IPDPS 2011 Charles Babbage Award. He is a Fellow of the AAAS, ACM, IEEE, and SIAM and a member of the National Academy of Engineering.

Rafael Ferreira da Silva

CNRS Software Engineer

PhD candidate at INSA-Lyon

CREATIS

www.rafaelsilva.com

Geoffrey Fox

Institute: Indiana University

Fox received a Ph.D. in Theoretical Physics from Cambridge University and is now distinguished professor of Informatics and Computing, and Physics at Indiana University.  He currently works in applying computer science to Bioinformatics, Earthquake and Ice-sheet Science, and still works in Particle Physics.  He is principal investigator of FutureGrid – a facility to enable development of new approaches to computing. He is a Fellow of ACM and APS.

Al Geist

Institute: ORNL

Al Geist is a Corporate Research Fellow at Oak Ridge National Laboratory. He is the Chief Technology Officer of the Leadership Computing Facility His recent research is on Exascale computing and resilience needs of the hardware and software.  In his 28 years at ORNL, he has published two books and over 200 papers in areas ranging from heterogeneous distributed computing, numerical linear algebra, parallel computing, collaboration technologies, solar energy, materials science, biology, and solid state physics.  You can find out more about Al and a complete list of his publications on his web page: http://www.csm.ornl.gov/~geist

Andrew Grimshaw

Institute: U Virginia

Andrew Grimshaw received his Ph.D. from the University of Illinois at Urbana-Champaign in 1988. He joined the University of Virginia as an Assistant Professor of Computer Science, becoming Associate Professor in 1994 and Professor in 1999. He is the chief designer and architect of Mentat, Legion, Genesis II, and the co-architect for XSEDE. In 1999 he co-founded Avaki Corporation, and served as its Chairman and Chief Technical Officer until 2003. In 2003 he won the Frost and Sullivan Technology Innovation Award. In 2008 he became the founding director of the University of Virginia Alliance for Computational Science and Engineering (UVACSE). The mission of UVACSE is to change the culture of computation at the University of Virginia and to accelerate computationally oriented research.

Andrew is a leading member of the Open Grid Forum (OGF), serving both as a member of the OGF's Board of Directors and as Architecture Area Director.  Andrew is the author or co-author of over 50 publications and book chapters.  His current projects are Genesis II and XSEDE. Genesis II, is an open source, standards-based, Grid system that focuses on making Grids easy-to-use and accessible to non computer-scientists. XSEDE (eXtreme Science and Engineering Discovery Environment) is the NSF follow-on to the TeraGrid project.

Bill  Gropp

Institute: UIUC

William Gropp received his B.S. in Mathematics from Case Western Reserve University in 1977, a MS in Physics from the University of Washington in 1978, and a Ph.D. in Computer Science from Stanford in 1982. He held the positions of assistant (1982-1988) and associate (1988-1990) professor in the Computer Science Department at Yale University. In 1990, he joined the Numerical Analysis group at Argonne, where he was a Senior Computer Scientist in the Mathematics and Computer Science Division, a Senior Scientist in the Department of Computer Science at the University of Chicago, and a Senior Fellow in the Argonne-Chicago Computation Institute. From 2000 through 2006, he was also Deputy Director of the Mathematics and Computer Science Division at Argonne. In 2007, he joined the University of Illinois at Urbana-Champaign as the Paul and Cynthia Saylor Professor in the Department of Computer Science. In 2008, he was appointed Deputy Director for Research for the Institute of Advanced Computing Applications and Technologies at the University of Illinois. In 2011, he became the founding Director of the Parallel Computing Institute. His research interests are in parallel computing, software for scientific computing, and numerical methods for partial differential equations. He has played a major role in the development of the MPI message-passing standard. He is co-author of the most widely used implementation of MPI, MPICH, and was involved in the MPI Forum as a chapter author for both MPI-1 and MPI-2. He has written many books and papers on MPI including "Using MPI" and "Using MPI-2". He is also one of the designers of the PETSc parallel numerical library, and has developed efficient and scalable parallel algorithms for the solution of linear and nonlinear equations. Gropp is a Fellow of ACM, IEEE, and SIAM and received the Sidney Fernbach Award from the IEEE Computer Society in 2008. Gropp is a member of the National Academy of Engineering.

Torsten Hoefler

Institute: ETH, Zurich

Torsten is an Assistant Professor of Computer Science at ETH Zürich, Switzerland.  Before joining ETH, he lead the performance modeling and simulation efforts of parallel petascale applications for the NSF-funded Blue Waters project.  He is also a key member of the Message Passing Interface (MPI) Forum where he chairs the "Collective Operations and Topologies" working group.  Torsten received his Ph.D. in Computer Science from Indiana University.  He won the best paper award at the ACM/IEEE Supercomputing Conference 2010 (SC10), published over 40 peer-reviewed scientific conference and journal articles and authored chapters of the MPI-2.2 and MPI-3.0 standards.  Torsten received the SIAM SIAG/Supercomputing Junior Scientist Prize in 2012.  His research interests revolve around the central topic of "Performance-centric Software Development" and deal with scalable networks, parallel programming techniques, and performance modeling.  Additional information about Torsten can be found on his homepage at unixer.de.

Jeff Hollingsworth

Institute: U Maryland

Jeffrey K. Hollingsworth is a Professor of the Computer Science Department at the University of Maryland, College Park. He received his PhD and MS degrees in computer sciences from the University of Wisconsin.  He received a B. S. in Electrical Engineering from the University of California at Berkeley.  Dr. Hollingsworth’s research seeks to develop a unified framework to understand and improve the performance of large systems.  He is Editor in chief of the journal Parallel Computing, and is the general chair of the SC12 conference

Emmanuel Jeannot

Institute: INRIA

Emmanuel Jeannot is a research scientist at INRIA (Institut National de Recherche en Informatique et en Automatique) and he is doing his research at INRIA Bordeaux Sud-Ouest and at the LaBRI laboratory since Sept. 2009. Before that he held the same position at INRIA Nancy Grand-Est. From Jan. 2006 to Jul. 2006, he was a visiting researcher at the University of Tennessee, ICL laboratory. From Sept. 1999 to Sept. 2005, he was assistant professor at the Université Henry Poincaré, Nancy 1. During the period of 2000 to 2009, he did his research at the LORIA laboratory. He got his PhD and Master degree in computer science (resp., in 1996 and 1999) both from Ecole Normale Supérieur de Lyon, at the LIP laboratory. After his PhD, he spent one year as a postdoc at the LaBRI laboratory in Bordeaux. His main research interests are scheduling for heterogeneous environments and grids, data redistribution, algorithms and models for parallel machines, grid computing software, adaptive online compression and programming models.

Thilo Kielmann

Institute: Vrije Universiteit, Amsterdam

Thilo Kielmann studied Computer Science at Darmstadt University of Technology, Germany. He received his Ph.D. in Computer Engineering in 1997, and his habilitation in Computer Science in 2001, both from Siegen University, Germany. Since 1998, he is working at Vrije Universiteit Amsterdam, The Netherlands, where he is currently Associate Professor at the Computer Science Department.  His research interests are in the area of high-performance, distributed computing, namely programming environments and runtime systems for applications on clusters, grids, and clouds.

Laurent Lefevre

Institute: INRIA

Dr Laurent Lefèvre obtained his Ph.D. in Computer Science in January 1997 at LIP Laboratory (Laboratoire Informatique du Parallélisme) in ENS-Lyon (Ecole Normale Supérieure), France. From 1997 to 2001, he was assistant professor in computer science in Lyon 1 University and a member of the RESAM Laboratory (High Performance Networks and Multimedia Application Support Lab.) Since 2001, he is a permanent researcher in computer science at INRIA (the French Institute for Research in Computer Science and Control). He is a member of the RESO team (High Performance Networks, Protocols and Services) from the LIP laboratory in Lyon, France. He has organized several conferences in high performance networking and computing and he is a member of several program committees. He has co-authored more than 80 papers published in refereed journals and conference proceedings. He is a member of IEEE and takes part in several research projects. His research interests include: distributed computing and networking, Green and Energy Efficient Computing and Networking, autonomic networking, high performance networks protocols and services.

Andrew Lumsdaine

Institute: Indiana U

Andrew Lumsdaine is a Professor of Computer Science in the School of Informatics & Computing at Indiana University and Director of the Center for Research in Extreme Scale Technologies (CREST) at the Pervasive Technology Institute.  Lumsdaine received his Ph.D. from MIT in 1992.  Prior to coming to IU, he was a faculty member in the Department of Computer Science and Engineering at the University of Notre Dame (1992-2001). His research interests include high[-performance computing, computational science and engineering, parallel and distributed programming, and computational photography.  Lumsdaine is a member of ACM, IEEE, and SIAM, as well as the MPI Forum, and the ISO C++ standards committee.

Rusty Lusk

Institute: Argonne

Ewing “Rusty” Lusk is an Argonne Distinguished Fellow Emeritus at Argonne National Laboratory.  He received his B.A. in mathematics from the University of Notre Dame in 1965 and his Ph.D. in mathematics from the University of Maryland in 1970.  He was a professor of mathematics and computer science at Northern Illinois University before joining Argonne in 1982.  He was the director of the Mathematics and Computer Science Division at Argonne from 2006 to 2011.  His research interests include programming models for scalable parallel computing, implementation issues for the MPI Message-Passing Interface standard, parallel performance analysis tools, and system software for large-scale machines.  He is the author of five books and more than a hundred research articles in mathematics, automated deduction, and parallel computing.  Current activities include co-directorships of the DOE Co-Design Center for Exascale Simulation of Advanced Reactors (CESAR) and the SciDAC-3 Project Nuclear Computational Low Energy Initiative (NUCLEI) in computational nuclear physics.

Satoshi Matsuoka

Institute: Tokyo Institute of Technology

Satoshi Matsuoka is a Professor at the Global Scientific Information and Computing Center of Tokyo Institute of Technology (GSIC). He is the leader of TSUBAME series of supercomputers, which became the 4th fastest in the world on the Top500 and awarded the "Greenest Production Supercomputer in the World" by the Green 500 in November, 2010 and June 2011. He has also co-lead the Japanese national grid project NAREGI during 2003-2007, and is currently leading various projects such as the JST-CREST Ultra Low Power HPC　and JSPS Billion-Scale Supercomputer Resilience.  He has authored over 500 papers according to Google Scholar, and has chaired many ACM/IEEE conferences, including the Technical Papers Chair, Community Chair, and the upcoming Program Chair for Supercomputing Conferences 09, 11 and 13 respectively.  He is a fellow of ACM and European ISC, and has won many awards including the JSPS Prize from the Japan Society for Promotion of Science in 2006, awarded by his Highness Prince Akishinomiya, the ACM Gordon Bell Prizes for 2011, and the Commendation for Science and Technology by the Minister of Education, Culture, Sports, Science and Technology in 2012.

Doug Miles

Institute: The Portland Group, Inc.

Douglas Miles is responsible for all business and technical operations of The Portland Group (PGI).  He has worked over the last 25 years in various positions in HPC applications engineering, math library development, technical marketing and management at FPS, Cray Research Superservers, PGI and STMicroelectronics.

Dr.-Ing. Bernd Mohr

Institute: Juelich

Senior Scientist, Jülich Supercomputing Center

Dr.-Ing. Bernd Mohr started to design and develop tools for performance analysis of parallel programs already with his diploma thesis at the University of Erlangen in Germany in 1987, and continued this in his Ph.D. work. During a three year Post-doc position at the University of Oregon, he was responsible for the design and implementation of the original TAU performance analysis framework for the parallel programming language pC++. Since 1996 he is a senior scientist at the Research Center Jülich Starting 2007, he became deputy head of the division "Application Support" where he is in charge of user support and training in regard to programming and performance tools at the Juelich Supercomputing Centre (JSC). In his research, he is leading the KOJAK and Scalasca performance tools efforts together with Felix Wolf of GRS in Aachen. He is active member in the International Exascale Software Project (IESP) and work package leader in the European (EESI) and Juelich (EIC, ECL) Exascale efforts. For the SC Conference series, he serves on the Steering Committee. He is the author of several dozen conference and journal articles about performance analysis and tuning of parallel programs.

http://www.fz-juelich.de/SharedDocs/Personen/IAS/JSC/EN/staff/mohr_b.html

Raymond Namyst

Institute: U Bordeaux & INRIA

Raymond Namyst is a Professor at University of Bordeaux, France. He is the scientific leader of the "Runtime" INRIA Research Group (http://runtime.bordeaux.inria.fr) devoted to the design of high performance runtime systems for parallel architectures.  His main research interests are in parallel computing, scheduling on heterogeneous multiprocessor architectures (multicore, NUMA, GPU), and communications over high speed networks. He has played a major role in the development of many significant software (ForestGOMP, MPICH/Nemesis, etc.), and most notably the StarPU runtime system (http://runtime.bordeaux.inria.fr/StarPU/). 

Mike Norman

Institute: UCSD

Michael L. Norman is Director of the San Diego Supercomputer Center and Distinguished Professor of Physics at UCSD where he also directs the Laboratory for Computational Astrophysics. He is the Principal Investigator of the $20M Gordon data-intensive supercomputer project which seeks to bring high performance computing to “big data” applications. He received his B.S. from Caltech in 1975, and his Ph. D. from UC Davis in 1980. After holding appointments at the Lawrence Livermore and Los Alamos National Laboratories, the Max Planck Institute for Astrophysics, and the National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign where he was also Professor of Astronomy, he joined the faculty at UCSD in 2000. His research focus is the computer simulation of astronomical phenomena using supercomputers, and the development of the numerical methods to carry them out. He is the author of over 250 papers on diverse topics including star formation, cosmic jets, and cosmological evolution. His lab pioneered the concept of community application software for astrophysical and cosmological simulations through the widely used ZEUS and ENZO codes. His computer visualizations have appeared in numerous educational TV shows and films, including PBS Nova and The Discovery Channel. He is the recipient of Germany’s Alexander von Humboldt Research Prize and the IEEE Sidney Fernbach Award. He was elected Fellow of the American Physical Society in 2001, and the American Academy of Arts and Sciences in 2005.

Christian Obrecht

Institute: EDF R&D / CETHIL UMR5008

Christian Obrecht is currently working as an engineer for EDF R&D and pursuing doctoral studies at the CETHIL laboratory in INSA-Lyon under the supervision of B. Tourancheau (LIG), J.-J. Roux and F. Kuznik (CETHIL). He graduated in mathematics from ULP-Strasbourg I in 1990 and taught at high school level until 2008. He received a MSc degree in computer science from UCB-Lyon I in 2009. His research work focuses on implementation and optimization strategies for parallel CFD applications on emerging many-core architectures.

Cherri Pancake

Institute: Oregon State

Phil  Papadopoulos

Institute: UC San Diego

Dr. Papadopoulos received his PhD in 1993 from UC Santa Barbara in Electrical Engineering. He spent 5 years at Oak Ridge National Laboratory as part of the the Parallel Virtual Machine (PVM) development team. He came to UCSD as research professor in computer science in 1998 and still holds and adjunct appointment.  He is currently the Chief Technology Officer at the San Diego Supercomputer Center (SDSC).  He is the architect  of the Triton resource  (tritonresource.sdsc.edu) which is a network, storage and memory rich, mid-scale (3000 core) cluster that is also an ongoing experiment in economic sustainability for campus-scale computing.   In addition to duties at SDSC,  his research interests revolve around distributed, clustered, and cloud-based systems and how they can be used more effectively in an  expanding bandwidth-rich environment.  Dr. Papadopoulos is a key investigator for key research projects at UCSD including the The National Biomedical Computation Resource(NBCR) and the Pacific Rim Applications and Grid Middlware Assembly (PRAGMA)  He is well known for leading the development of the open-source, NSF-funded Rocks Cluster toolkit (OCI-0721623), which has installed base of 1000s of clusters.   Rocks (www.rocksclusters.org) is used for both research and production systems with scalability to 1000s of nodes.  Our work in Rocks focuses on developing practical, scalable, and  robust virtual machine authoring and implementation of hybrid clusters that  consist of both real and virtual hardware. More recently,  we have demonstrated seamlessly extending the number of nodes in a  local cluster with additional virtualized resources housed on campus clouds and/or  in Amazon EC2.

Jelena Pjesivac-grbovic

Institute: Google

Jelena Pjesivac-Grbovic is a senior software engineer in Systems Infrastructure at Google, focusing on distributed data processing frameworks. She received PhD from the University of Tennessee, Knoxville under supervision of Dr. Jack J. Dongarra. Prior to joining Google she actively contributed to the implementation and optimizations of the MPI collective operations in OpenMPI and FT-MPI projects.

Padma Raghavan

Institute: Penn State

Padma Raghavan is a Distinguished Professor of Computer Science and Engineering at The Pennsylvania State University and the Director of the university-wide Institute for CyberScience for promoting interdisciplinary computation and data enabled research. Raghavan’s research concerns high-performance computing and its applications. She has over eighty peer-reviewed publications in the areas of parallel sparse scientific computing and energy-aware supercomputing, i.e., performance and energy scalability of advanced computer systems. Raghavan serves on the program committees of major conferences sponsored by ACM, IEEE and SIAM including co-chair of the SIAM Conference on Computational Science and Engineering in 2011 and co-chair of Technical Papers at IEEE/ACM Supercomputing in 2012.  Raghavan currently serves on various editorial and advisory boards including the SIAM series on Computational Science and Engineering, the SIAM Series on Software, Environments and Tools, the Journal of Parallel and Distributed Computing, the External Review Committee of the Computation Directorate of the Lawrence Livermore National Laboratory, and the National Academies Panel on Digitization and Communication Science.

Jonathan Rouzaud

Institute: Lyon

Joel Saltz

Institute: Emory U

Joel Saltz, MD, PhD: Dr. Saltz is Director of the Center for Comprehensive Informatics, Professor and Chair of Biomedical Informatics at Emory University. Over the course of his career, he has developed a rich set of middleware, optimization and runtime compilation methods targeting data and compute intensive irregular, adaptive and multi-resolution applications. He was the originator of the inspector-executor compilation framework and extended this framework to handle a wide variety of computationally intensive applications as well as the optimization of queries that target disk based multi-resolution datasets. The Saltz team has developed various techniques and tools (e.g. DataCutter and Active Disks) that support in-transit data processing. Dr. Saltz has also contributed to the development of ambitious biomedical algorithms and methods that target high end computers, very large scale storage systems and grid environments.

Vaidy Sunderam

Institute: Emory University

Vaidy Sunderam is a professor at Emory University.  His research interests are in parallel and distributed processing systems and infrastructures for collaborative computing. His prior and current research efforts have focused on system architectures and implementations for heterogeneous metacomputing, including the Parallel Virtual Machine system and several other frameworks such as IceT, CCF, Harness, and Unibus.

Frédéric Suter

Institute: IN2P3/CNRS Computing Center

Frédéric Suter is a CNRS junior researcher in Computer Science at the IN2P3 Computer Center in Lyon, France since October 2008. His research interests include scheduling, grid computing and platform and application simulation. He obtained his M.S. from the Université Jules Verne, Amiens, France in 1999 and his Ph.D. from the Ecole Normale Supérieure de Lyon, France in 2002. Prior to joining the CNRS, he was an assistant professor at Nancy University, France. He is currently involved in the SONGS (Simulation of Next Generation Systems) project based on the SimGrid toolkit (http://simgrid.inria.gforge.fr) and leader of the work package on Data Grids. He is particularly interesting in gathering and analyzing usage logs of storage systems ans IaaS Clouds. He was also mandated by the French NGI to foster collaborations between Computer Science researchers and operators of the production Grid on the topic of Cloud Computing.

Martin Swany

Institute: Indiana U

Rajeev Thakur

Institute: Argonne

Rajeev Thakur is the Deputy Director of the Mathematics and Computer Science Division at Argonne National Laboratory, where he is also a Senior Computer Scientist. He is also a Senior Fellow in the Computation Institute at the University of Chicago and an Adjunct Associate Professor in the Department of Electrical Engineering and Computer Science at Northwestern University. He received a Ph.D. in Computer Engineering from Syracuse University in 1995.  His research interests are in high-performance computing, parallel programming models, runtime systems, communication libraries, and parallel I/O.

Bernard Tourancheau

Institute: University Grenoble

Bernard Tourancheau got MSc in Apply Maths from Grenoble University in 1986 and a MSc in Renewable Energy Science and Technology from Loughborough University in 2007. He was awarded best Computer Science PhD by Institut National Polytechnique of Grenoble in 1989 for his work on Parallel Computing for Distributed Memory Architectures.

Working for the LIP laboratory, he was appointed assistant professor at Ecole Normale Supérieure de Lyon in 1989 before joining CNRS as a Senior Researcher. After initiating a CNRS-NSF collaboration, he worked two and half years on leave at the University of Tennessee on a senior researcher position with the US Center for Research in Parallel Computation at the ICL laboratory.

He then took a Professor position at University of Lyon in 1995 where he created a research laboratory and the INRIA EPI RESO, specialized in High Speed Networking and Clusters.

In 2001, he joined SUN Microsystems Laboratories for a 6 years sabbatical as a Principal Investigator in the DARPA HPCS project where he lead the backplane networking group.

Back in academia he oriented his research on sensor and actuator networks for building energy efficiency at the LIP and CITI labs associated with INRIA.

He was appointed Professor at University Joseph Fourier of Grenoble in 2012. Since then, he is developing research at the LIG laboratory Drakkar team about protocols and architectures for sensor networks and their applications to energy efficiency in buildings as well as GPGPU's algorithms optimization.

He has authored more than an hundred peer-reviewed publications and owns 8 patents.

Anne Trefethen

Institute: Oxford

Jeffrey S. Vetter

Institute: ORNL and Georgia Tech

Jeffrey Vetter, Ph.D., holds a joint appointment between Oak Ridge National Laboratory (ORNL) and the Georgia Institute of Technology (GT). At ORNL, Vetter is a Distinguished R&D Staff Member, and the founding group leader of the Future Technologies Group in the Computer Science and Mathematics Division. At GT, Vetter is a Joint Professor in the Computational Science and Engineering School of the College of Computing, the Principal Investigator and Project Director for the NSF Track 2D Experimental Computing Facility for large scale heterogeneous computing using graphics processors, and the Director of the NVIDIA CUDA Center of Excellence. Professionally, Vetter has published over 110 peer-reviewed papers; he has served on over 50 program committees, in leadership roles, including most recently as the SC12 Technical Papers Co-chair, and as an associate editor of IEEE Transactions on Computers. His papers have won awards at the International Parallel and Distributed Processing Symposium and EuroPar; he was awarded the ACM Gordon Bell Prize in 2010.

Xavier Vigouroux

Institute: Bull

Xavier is Based in Grenoble in France where he works for the Bull HPC business unit as HPC Business Develoment Manager for Education and Research Market. Prior to that, he has been Bull HPC Benchmark Manager for 5 years. He also worked on high availability, cluster, and, High Producitivy Computer Systems for Sun Microsystems and Sun labs, and HP. He obtained his PhD in 1996 with Bernard as advisor.

Marlisha Vucinic

Institute: LIG, Grenoble U

I enrolled Electrical Engineering at University of Montenegro in 2006. I was awarded a scholarship by USAID to spend the third year of my bachelor program at Tennessee Tech in Cookeville, TN, where I followed Electrical and Computer Engineering courses. Upon my return to Montenegro I finalized my undergraduate studies by obtaining a 4 year national engineering diploma. After this, I went for a double-degree Master called Communication Systems Engineering (first year of Master at Politecnico di Torino in Italy, the second at ENSIMAG, Grenoble, France). My research so far has been mostly focused to wireless networks from a protocol perspective (802.11 - MAC issues and IPv6 sensor networks - routing challenges) but I also have one publication from 2010 on digital design of an ethernet switch on a NetFPGA platform. I will be doing my Phd starting from cca October on Security issues in IPv6 sensor networks.

CCGSC 1998 Participants, Blackberry Tennessee

CCGSC 2000 Participants, Faverges, France

CCGSC 2002 Participants, Faverges, France

CCGCS 2004 Participants, Faverges, France

CCGCS 2006 Participants, Flat Rock North Carolina

Some additional pictures can be found here.

CCGCS 2008 Participants, Flat Rock North Carolina

CCGCS 2010 Participants, Flat Rock North Carolina

CCDSC 2012 Participants, Dareize, France