DAMP 2010: Workshop

Declarative Aspects of Multicore Programming

Madrid, SPAIN

(colocated with POPL 2010)

January 19, 2010

damp10.cs.nmsu.edu

SUBMISSION DEADLINE: SEPTEMBER 21, 2009

The  advent  of multicore architectures  has profoundly  increased the  importance of research  in parallel  computing. Modern  platforms  are becoming more complex and heterogenous and novel solutions  are needed to account for their peculiarities.

Multicore  architectures will  differ in  significant ways  from their multisocket  predecessors. For example,  the communication  to compute bandwidth ratio is  likely to be higher, which  will positively impact performance. More generally, multicore architectures introduce several new  dimensions  of variability  in  both  performance guarantees  and architectural  contracts,  such as  the  memory  model,  that may  not stabilize for several generations of product.

Programs  written  in  functional  or  (constraint-)logic  programming languages, or in other highly declarative languages  with a controlled   use of side effects, can  greatly simplify parallel  programming. Such declarative programming  allows  for  a  deterministic semantics  even  when  the underlying implementation might be highly non-deterministic.  In addition to simplifying programming this can simplify debugging and analyzing correctness.

DAMP 2010 is the  fifth in  a series of  one-day  workshops seeking to explore  ideas in  declarative  programming  language design that will greatly  simplify programming for  multicore  architectures, and  more generally  for tightly  coupled parallel  architectures.  The emphasis will  be  on (constraint-)logic and  functional  programming, but  any declarative programming language ideas that aim to raise the level of abstraction are welcome. DAMP seeks to gather together researchers in declarative approaches to parallel programming and to foster cross fertilization across different approaches.

Specific topics include, but are not limited to:

 * investigation  of applications  of logic,  constraint logic, and    functional programing to multicore programing

 * run-time issues of exploitation of parallelism using declarative    programming approaches (e.g., garbage collection, scheduling)

 * architectural   impact  on  exploitation  of  parallelism   from declarative languages

 * type systems and analysis for accurately detecting dependencies,  aliasing, side effects, and impure features

 * language level declarative constructs for expressing parallelism  

* declarative language  specification for  the description of data placement and distribution

 * compilation   and   static analysis   techniques   to    support exploitation  of  parallelism from declarative  languages (e.g.,

   granularity control)

 * practical  experiences  and  challenges  arising  from  parallel declarative programming

 * technology for debugging parallel programs

 * design  and   implementation   of  domain-specific   declarative languages for multicore programming

Submission:

  Submitted  papers  papers  should  not  exceed  10  pages  in ACM

  SIGPLAN conference format. Submission is electronic via:

  http://www.easychair.org/conferences/?conf=damp10

  Accepted papers will be published in the ACM Digital  Library and  in a physical  proceedings.  Papers must  adhere to  the  SIGPLAN Republication Policy:

       http://www.sigplan.org/republicationpolicy.htm

  Concurrent submissions to other conferences,  workshops, journals,   or similar forums of  publication are  not allowed.  However, DAMP  is   intended to  be  a  venue for  discussion and  exploration of works-in-progress, and so publication of a paper  at DAMP  2010 is not intended to preclude later publication as appropriate.

  Additional information  about the submission process can be  found at the conference web site.

Important dates:

  Abstract submission:     Sept. 21

  Paper submission:        Sept. 25

  Notification to authors: Oct.  26

  Camera ready:            Nov.   9

Program Chair:

  Enrico Pontelli

  New Mexico State University

General Chairs:

  Leaf Petersen

  Intel Corporation

  Santa Clara, CA, USA

Program Committee:

   Manuel Carro           Universidad Politecnica de Madrid

   Clemens Grelck         University of Hertfordshire

   Haifeng Guo            University of Nebraska at Omaha

   Gabriele Keller        University of New South Wales

   Hans-Wolfgang Loidl    Ludwig-Maximilians-Universitat Munchen

   Leaf Petersen          Intel Corporation

   John Reppy             University of Chicago

   Ricardo Rocha          University of Porto

   Kostis Sagonas         National Technical University of Athens

   Vitor Santos Costa     University of Porto

   Satnam Singh           Microsoft Research

   Philip Trinder         Heriot-Watt University

   Pascal Van Hentenryck  Brown University

URL:  http://damp10.cs.nmsu.edu

I'm interested too. Is there any place for one more ?