Sunday 4 March 2012. London.
A workshop co-located with ASPLOS 2012.
David F. Bacon IBM Research Satnam Singh The University of Birmingham
Compiling programs into circuits is becoming increasingly important as we strive to reduce the latency and energy consumption of our computing systems. Although significant effort has been put into the task of "C to gates" synthesis which aims to improve the productivity of hardware engineers there has been little work in the area of converting complete programs written by software engineers into circuits. The purpose of this workshop is to bring together researchers and developers working on novel techniques for translating programs into circuits as well as people using alternative techniques for translating algorithms into circuits without using conventional hardware description languages based on event-based simulation semantics. Our objective is to foster interaction between researchers and developers working in this area and to establish a community that can help drive forward a common set of goals and principles for the synthesis of programs into circuits.
Examples of topics of interest include:
System design using alternative hardware description languages like Bluespec.
Case-studies and experience reports e.g. line-speed packet processing from high level circuit descriptions.
Synthesis of algorithms in high level languages into circuits e.g. from C#, F#, Go, Java and Scala into hardware.
Analysis, verification and debugging of hardware systems described in high level languages (including formal verification techniques and language semantics).
Embedded domain specific languages for hardware design e.g. Lava.
Support for high-level language features e.g. dynamic recursion and dynamic data-types.
Application of concurrency and parallelism constructs in programming languages to hardware design (e.g. Cilk style programs, OpenCL/CUDA, LINQ etc.).
Numerical aspects e.g. specification and control of range and precision.
We have a limited number of presentation slots so we welcome suggestions for people to give talks about their experience of using or developing alternative techniques for implementing computations on circuits although we may not be able to fit in all the proposals. There is no plan to have a formal proceedings.
The article Computing Without Processors in the August 2011 edition of the Communications of the ACM gives one perspective on how programs may be turned into circuits for heterogeneous execution environments (there is also an ACM Queue version of the article Computing Without Processors).
If you would like to present your own work on this topic please contact one of the organizers: email@example.com or firstname.lastname@example.org
|Submissions:||January 9, 2012|
|Event:||March 4, 2012-March 4, 2012|