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 and call for abstracts
The workshop agenda includes a CnC tutorial on current
and future trends and talks selected from contributed
abstracts. 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 the
workshop chairs at
cncworkshop2016@gmail.com
no later than Monday, August 29, 2016. Please include:
- Name
- Affiliation
- At most one page abstract of the work to be presented
.
Location
The workshop will be held at Hilton Garden Inn, in the college town next to University of Rochester river campus (arts, sciences and engineering) and medical school.
Hotel Information
Accomodations information for CnC 2016 participants can be found here.
Workshop Agenda
Tuesday, September 27, 2016 (Salon A):
08:30 --
Workshop Registration
09:00 --
Welcome to CnC 2016
09:15 --
CnC Tutorial - Frank Schlimbach (Intel)
(slides)
10:00 -- Session 1: Principles
- Overwriting, Non-Deterministic and Safe Data-Puts in (Intel®) CnC - Frank Schlimbach (Intel) and Kath Knobe (Rice University).
(slides)
- CnC is a Dependence Programming Model - Kath Knobe (Rice University) and Zoran Budimlic (Rice University).
(slides; pptx)
- Declarative Communication for CnC - Zoran Budimlic (Rice University) and Kath Knobe (Rice University).
(pptx)
11:30 --
Networking and Lunch break (Salon B)
13:30 -- Session 2: Graph aspects in CnC
- Hierarchical CnC - Kath Knobe (Rice University).
(slides)
- Algorithmic Generation and Evaluation of Step-code Hierarchies in CnC Applications - Nick Vrvilo (Rice University). (slides)
- Demand-driven execution in CnC: First Steps - Peter Elmers (Rice University) and Nick Vrvilo (Rice University). (slides)
- Abstract graph semantics for CnC - Tiago Cogumbreiro (Rice University) (slides)
15:30 --
Afternoon Break (Salon A)
16:00 -- Session 3: Distributed
- PIPES: A Language and Compiler for Distributed Task based Programming - Martin Kong (Rice University), Louis-Noel Pouchet (Colorado State University), P. Sadayappan (Ohio State University), and Vivek Sarkar (Rice University).
(slides)
- On Improving the Execution of Distributed CnC Programs - Yuhan Peng (Rice University), Martin Kong (Rice University), Louis-Noel Pouchet (Colorado State University), and Vivek Sarkar (Rice University).
(slides)
- Towards Resilient CnC-OCR - Sara Hamouda (The Australian National University), Sanjay Chatterjee (Rice University), Nick Vrvilo (Rice University), Zoran Budimlic (Rice University), and Vivek Sarkar (Rice University).
(slides)
17:30 --
Networking
18:00 --
Dinner outing
Wednesday, September 28, 2016 (Salon A):
08:30 --
Continental Breakfast
09:00 --
Keynote Presentation: "Parallel Computation Models for HPC and Big Data Systems
-- From Dataflow to Codelets, and Beyond" - Guang R. Gao (University of Delaware)
10:00 -- Session 4: New directions and tuning in CnC
- Dynamic Task Speculation Support Through Divide-and-merge Memory Allocation - Chen Ding (University of Rochester), Benjamin O’Halloran (University of Rochester), Jacob Bisnett (University of Rochester), Joel Kottas (University of Rochester), and Colin Pronovost (University of Rochester)
(slides)
- Evaluating Performance of Task and Data Coarsening in Concurrent Collections - Chenyang Liu (Purdue University) and Milind Kulkarni (Purdue University).
(slides)
- Adaptive Tuning of Parallel Programs with CnC - Murali Emani (Lawrence Livermore National Laboratory).
(slides)
- Experimenting With Bringing Data-Analytics Kernels To CnC - Frank Schlimbach (Intel) and Kath Knobe (Rice University).
12:00 --
Closing remarks / End of CnC 2016
13:00 --
LCPC Keynote Presentation: "The Multi-core Problem as an Algorithmic Problem" - Leslie Valiant (Harvard University)
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
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.
Need more information?
If you have any questions about logistics or participation,
please contact the workshop chairs
at
cncworkshop2016@gmail.com.