The annual Concurrent Collections (CnC) workshop is as 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 with an interest programming models
inspired by dataflow and/or tuple space ideas as well as
current or emerging applications of such models.
Participation
There is no registration fee. To register, please send an email to james.brodman@intel.com and sadayappan.1@osu.edu
with the following information:
- Name
- Affiliation
- Background with CnC
- Dietary Information (for Boxed Lunches)
Hotel Information
The workshop has reserved a block of rooms at the Hampton Inn Champaign/Urbana at a special workshop rate. The Hampton Inn is located two blocks from the Siebel Center.
Link to Room Block
Agenda
The workshop agenda includes research and experience
presentations, a keynote address, a tutorial and
plenty of time will be left open for unstructured mixing,
mingling, and networking.
Thursday, December 6
- 08:15 -- Breakfast and Coffee
- 08:30 -- CnC Tutorial: Frank Schlimbach
- 10:00 -- Break
- 10:30 -- Tutorial, Continued
- 12:00 -- Lunch
- 13:00 -- Keynote: "Expressing Concurrency within and across Objects in Charm++" -- Professor Sanjay Kale
- 14:00 -- Tutorial, CnC Research Efforts: Zoran Budimlic
- 15:00 -- Break
- 15:30 -- Session 1: Applications
-
Parallelizing Compressive Sensing MRI via CnC-Babel and Matlab --
Richard Baraniuk, Zoran Budimlic, Michael Burke,
Shams Imam, Kath Knobe, Vivek Sarkar, Jianing Shi
-
BFS Preconditioning for High Locality, Data Parallel Algorithms --
Nicolas Vasilache, Benoit Meister, Muthu Baskaran, Richard Lethin
-
A Performance Evaluation of Distributed CnC on RTM --
Marc Tchiboukdjian, Bettina Krammer, Frank Schlimbach, Kath Knobe,
James Brodman
-
Overhead of CnC over TBB for Matrix Inversion --
Peiyi Tang
- 17:30 -- End Day 1
Friday, December 7
- 08:15 -- Breakfast and Coffee
- 08:30 -- Session 2: Programming Models
-
CDSC-GL: A CnC-inspired Graph Language for Mapping Applications onto Heterogeneous Devices -- Zoran Budimlic (Rice),
Louis-Noel Pouchet, Alina Sbirlea, Peng Zhang, Jason Cong,
Vivek Sarkar
-
Thoughts on the Intel Threading Building Blocks Flow Graph and Intel Concurrent Collections --
Mike Voss
-
Flexible Preconditions: A Model for Efficient Macro-Dataflow Execution --
Dragos Sbirlea, Alina Sbirlea, Kyle Wheeler, Vivek Sarkar
-
S-Net: A Coordination Language for Concurrent Collections --
A. Shafarenko and R. Kirner
- 10:30 -- Break
- 11:00 -- Session 3: Memory
-
Item Collections for Dense N-Dimensional Data Arrays in CnC --
Juergen Ributzka
-
Provably Correct Space-folding of CnC Programs -
Aaron W. Hsu
-
On the Scalability of Loop Tiling Techniques --
David G. Wonnacott, Michelle Mills Strout
- 12:30 -- Lunch
- 13:30 -- Session 4: Platforms
-
Mapping a Data-flow Programming Model onto Heterogeneous Platforms --
Alina Sbirlea, Yi Zou, Zoran Budimlic, Jason Cong, Vivek Sarkar
-
Execution Frontiers as Checkpoints in CnC --
Nick Vrvilo, Vivek Sarkar, Kath Knobe
-
Execution Frontiers in CnC: Increasing Execution Flexibility for Exascale Systems --
Kath Knobe
-
Early Experience with a CnC Tuning Capability --
Sanjay Chatterjee, Zoran Budimlic, Mike Burke, Vivek Sarkar,
Kath Knobe
- 15:30 -- End Day 2
Background on CnC
CnC is a parallel programming model for mainstream
programmers that differs from other approaches in its
philosophy. A CnC programmer doesn't specify parallel
operations; instead, he/she only specifies semantic ordering
constraints. This provides a separation of concerns between
the domain expert and the tuning expert, simplifying the job
of the domain expert while providing more flexibility to the
tuning expert. Details on CnC and related research can be
found at:
http://intel.ly/concurrent-collections
and
http://habanero.rice.edu/cnc
Prior workshops have served as a forum for users and potential
users of Concurrent Collections (CnC), to discuss experiences
with CnC and a range of topics, including developments for the
language, applications, usability, performance, semantics, and
teaching of CnC.
Sponsored by