<html><head><meta http-equiv="Content-Type" content="text/html charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">============================================================================<br> Workshop on Domain Specific Languages Design and Implementation (DSLDI)<br><br> Collocated with ECOOP 2013<br><br> Monday, July 1st, 2013, Montpellier, France<br><br> <a href="http://dsldi2013.hyperdsls.org/">http://dsldi2013.hyperdsls.org/</a><br>============================================================================<br><br><br>Modern hardware is growing more and more complex, often featuring not only<br>multiple cores but also heterogeneous components with various types of<br>architecturally different accelerators. Consequently, it is increasingly more<br>difficult for the programmers to produce high-performance scalable software,<br>which is often equally complex, using general-purpose programming languages<br>such as Java or C++, as they lack appropriate language-level abstractions.<br>Languages designed to support high productivity, such as scripting languages<br>exemplified by Python, JavaScript or Perl, make the programmer's task much<br>easier. Their performance, however, while certainly adequate for some use<br>cases, is not quite on-par with that of the general-purpose programming<br>languages. Domain Specific Languages (DSLs) combine the best features of the<br>general-purpose programming languages, that is efficiency, and of the<br>languages designed for high productivity, that is ease of programming. This<br>makes DSLs our best hope for harnessing computational resources available on<br>modern architectures without requiring super-human programming skills.<br><br>The goal of the DSLDI workshop is to bring together researchers and<br>practitioners interested in sharing ideas on how Domain Specific Languages<br>should be designed and implemented and on usage scenarios for modern DSLs. We<br>are interested both in discovering how already known domains, such as graph<br>processing or machine learning, can be best supported by DSLs but also in<br>exploring new domains that could be targeted by DSLs. More generally, we are<br>interested in building a community that can drive forward development of<br>modern DSLs.<br><br>The workshop will consist of a series of short invited talks whose main goal<br>would be to trigger exchange of opinion and discussions on the topics within<br>DSLDI's area of interest which include but are not limited to the following:<br><br>- DSL implementation techniques, including compiler-level and<br> runtime-level solutions<br>- utilization of domain knowledge for driving optimizations of DSL<br> implementations<br>- utilizing DSLs for managing parallelism and hardware heterogeneity<br>- DSL performance and scalability studies<br>- DSL tools, such as DSL editors and editor plugins, debuggers,<br> refactoring tools, etc.<br>- applications of DSLs to existing as well as emerging domains, for<br> example graph processing, image processing, machine learning,<br> analytics, robotics, etc.<br>- practitioners reports, for example descriptions of DSL deployment in<br> a real-life production setting<br><br>The workshop will be informal and will not have proceedings of any kind. We<br>have a limited number of presentation slots so we welcome suggestions for<br>people to give talks about their experience of using or developing DSLs. If<br>you would like to give a presentation then please contact Adam Welc at<br><a href="mailto:adam.welc@oracle.com">adam.welc@oracle.com</a> (presentation submission deadline: April 26th, 2013).<br><br><br>Organizing committee:<br><br>Hassan Chafi, Oracle Labs<br>Tim Harris, Oracle Labs<br>Kunle Olukotun, Stanford University<br>Tiark Rompf, EPFL<br>Satnam Singh, Google<br>Laurence Tratt, King's College London<br>Eelco Visser, Delft University of Technology<br>Adam Welc, Oracle Labs</body></html>