CnC-2014: The Sixth Annual Concurrent Collections Workshop
September 18-19, 2014 at Intel Corporation, 2111 NE 25th Ave., Hillsboro, Oregon 97124 (collocated with LCPC'14)
The annual Concurrent Collections (CnC) workshop is a forum for researchers and developers of parallel programs to interact on a variety of issues related to next-generation parallel programming models. The focus is on fostering a community around the CnC programming model. However, we also strongly encourage participation by anyone interested in programming models inspired by dataflow and/or tuple space ideas, as well as current or emerging applications of these models.
Registration for CnC'14 is mandatory for all participants. To register please visit: http://engrevent.tamu.edu/event/101113
Choose the "CnC-2014 Registration" option. All registrants will need to provide a $30 payment onsite (cash or check accepted).
Registration deadline is Thursday, September 18, 2014.
Click here to see the CnC 2014 Workshop Agenda.
Accommodations for CnC-2014 attendees will be provided by: Larkspur Landing3133 NE Shute RoadHillsboro, OR 97124Phone: (503) 681-2121
The room rate is $109 and is valid for September 17 and 18, 2014.
When making reservations, please refer to "CNC Workshop 2014." The deadline for reservations at the $109 meeting rate is Friday, September 12, 2014.
Participation and Call for Abstracts
The workshop agenda will include CnC tutorials on current and future trends, talks selected from contributed abstracts, and a programming contest. Topics of interest include, but are not limited to: language design and implementation, semantics and theory, application experiences, and teaching CnC.
If you are interested in giving a talk, please submit a short abstract (between 200–500 words in length) to workshop chairs Sanjay Chatterjee and John Feo no later than Friday, August 8, 2014.
This year, for the first time, we are challenging small teams (as small as one contestant) of CnC developers to compete in a programming contest. If you are interested in competing, please send a team name, list of team member(s), and a problem of choice by Friday, August 8, 2014 to workshop chairs Sanjay Chatterjee and John Feo.
We encourage you to work with the workshop committee to define an appropriate problem. Solutions will be judged on creativity, programming style, solution, and performance. A written report, including source code, is due to both workshop chairs (Chatterjee and Feo) by Friday, September 12, 2014.
We will reserve a session on the first day of the workshop for teams to present their work. While presenting is not required, it is highly encouraged as it provides an opportunity for the committee and audience to better understand your work. A significant prize will be awarded to the winning team at the close of the workshop.
Thursday, September 18, 2014 (JFCC 119):
08:30 -- Breakfast and Coffee
09:00 -- Intel Concurrent Collections Tutorial – James Brodman (Intel) [slides]
10:00 -- Morning Break
10:30 -- Keynote address – Intel’s Extreme-scale Computing Efforts – Bill Feiereisen (Intel) [slides]
11:30 -- State of CnC – Frank Schlimbach (Intel) [slides] and Zoran Budimlic (Rice University) [slides]
12:30 -- Lunch Break
13:30 -- Session 1: Applications
- Using CnC to Enable Re-use in the Portland Observatory – Kristin Tufte (Portland State University), Basem Elazzabi (Portland State University), Veronika Megler (Portland State University), Morgan Harvey (Portland State University), Kath Knobe (Rice University), and David Maier (Portland State University) [slides]
- Adaptive Mesh Refinement under the Concurrent Collections Programming Model – Robert Pavel (Los Alamos National Laboratory/University of Delaware), Robert Bird (Los Alamos National Laboratory/University of Warwick), Pascal Grosset (Los Alamos National Laboratory/University of Utah), Ken Czuprynski (Los Alamos National Laboratory/University of Iowa), Andrew Reisner (Los Alamos National Laboratory/University of Illinois at Urbana-Champaign), Erin Carrier (Los Alamos National Laboratory/University of Illinois at Urbana-Champaign), Christoph Junghans (Los Alamos National Laboratory), Benjamin Bergen (Los Alamos National Laboratory), and Allen McPherson (Los Alamos National Laboratory) [slides]
- Gene Sequencing Without The Pain – Yves Vandriessche (Vrije Universiteit Brussel) [slides]
- Experience Porting LULESH to CnC – Ellen Porter (Pacific Northwest National Laboratory) [slides]
15:30 -- Afternoon Break
16:00 -- Session 2: Runtime
- Statistic Multiplexed Computing – A Paradigm Shift Towards Data Parallel Tuple Space Computing – Justin Y. Shi (Temple University) [slides]
- CnC in an Event-driven Programming Model – Nick Vrvilo (Rice University), Brian Nickerson (Intel), Kath Knobe (Rice University), and Vivek Sarkar (Rice University) [slides]
- DFGR: an Intermediate Graph Representation for Macro-Dataflow Programs – Alina Sbirlea (Rice University), Louis-Noel Pouchet (UCLA), and Vivek Sarkar (Rice University) [slides]
17:30 -- Networking
18:30 -- Dinner outing
Friday, September 19, 2014 (JFCC 117):
08:30 -- Breakfast and Coffee
09:00 -- Future of CnC – Kathleen Knobe (Rice University) [slides]
10:00 -- Morning Break
10:30 -- Session 3: Programming Contest Talks
- Parsec Apps – SeMiBo (Sergei Krestianskov, Michael Buse, and Bo Zhao, German Research School for Simulation Sciences) [slides]
- Implementation of Classical Molecular Dynamics Algorithm Using Intel Concurrent Collections – Exmatex Co-design Summer School Team (Riyaz Haque, Lawrence Livermore National Laboratory) [slides]
- Fast Linear Algebra Kernels with CnC – Buckeyes (Martin R. Kong and Louis-Noel Pouchet, Ohio State University) [slides]
12:00 -- Lunch Break
13:00 -- Session 4: Programming Contest Talks
- CnC for Ray Tracing – WSU Graphics (Ellen Porter and Robert R. Lewis, Pacific Northwest National Laboratory) [slides]
- Jacobi2d (Iterative Convergence, Stencil Computation) – Rice Owls (Shams Imam and Nick Vrvilo, Rice University) [slides]
14:00 -- Open Discussion/Panel Discussions
15:00 -- Afternoon Break
15:30 – Programming Contest Winner Announced – John Feo
16:00 -- End of CnC 2014
Background on CnC
CnC is a parallel programming model for mainstream programmers that philosophically differs from other approaches. CnC programmers do not specify parallel operations. Instead, they only specify semantic ordering constraints. This provides a separation of concerns between the domain expert and tuning expert, simplifying the domain expert’s job while providing more flexibility to the tuning expert.
Details on CnC and related research can be
Prior CnC workshops have served as a forum for users and potential users to discuss experiences with the programming, as well as a range of topics, including developments for the language, applications, usability, performance, semantics, and teaching CnC.