BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium : Correct-by-Construction Multiprocessor Programming
DTSTART:20140929T161500
DTEND:20140929T173000
DTSTAMP:20260407T110911Z
UID:0028fc2ed96705c6efebfed112beaa010c6af80c20e6fe6c98c9524d
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Albert Cohen - INRIAVideo of his talkAbstract:\nThis talk 
 is for people who care about program correctness and performance. About ac
 hieving both without resorting to impossible verification problems and tar
 get-specific optimizations. We will explore two complementary attempts to 
 regain control of your favorite multi- or many-core system:\n- streaming l
 anguages (with task-parallel runtimes)\;\n- polyhedral compilation (for DS
 Ls\, library generation\, portability).\nStream computing is often associa
 ted with regular\, data-intensive applications\, and more specifically wit
 h the family of cyclo-static data-flow models. The term also refers to bul
 k-synchronous data parallelism on SIMD and GPU architectures. Both interpr
 etations are valid but incomplete: streams underline the formal definition
  of Kahn process networks\, a foundation for deterministic concurrent lang
 uages and systems with a solid heritage. Stream computing is a semantical 
 framework for parallel languages and a model for pipelined\, task-parallel
  execution. Supporting research on parallel languages with dynamic\, neste
 d task creation and first-class streams\, we propose a new lock-free algor
 ithm for stalling and waking-up tasks in a user-space scheduler according 
 to changes in the state of the corresponding queues. The algorithm is port
 able and proven correct against the C11 memory model. We show through expe
 riments that it can serve as a keystone to efficient parallel runtime syst
 ems.\nCompilers face a never ending race to provide performance portabilit
 y over a moving computer architecture target. And the compilation problem 
 itself is also changing: programming becomes increasingly synonymous with 
 concurrent or parallel programming\, and beyond portability\, the need to 
 generate highly optimized\, resource-efficient code for common computation
 al tasks is also rejuvenating compiler construction. The polyhedral model 
 of compilation is a powerful framework addressing both applications. Progr
 ams are represented using systems of affine (linear) inequalities\, allowi
 ng to construct and search for advanced loop optimizations. We will study 
 ongoing work to extend the reach of the framework\, from dynamic\, data-de
 pendent control flow\, to the support of domain-specific languages and act
 ive libraries. And we will survey a few challenges to the adoption of poly
 hedral techniques in production compilers.Bio:\nAlbert Cohen is a senior r
 esearch scientist at INRIA and a part-time associate professor at École P
 olytechnique\, Paris\, France. He graduated from École Normale Supérieur
 e de Lyon\, and received his PhD from the University of Versailles in 1999
  (awarded two national prizes). He has been a visiting scholar at the Univ
 ersity of Illinois in 2000 and 2001\, and an invited professor at Philips 
 Research (then NXP Semiconductors)\, Eindhoven in 2006 and 2007.\nAlbert w
 orks on optimizing compilers for high-performance and embedded systems\, a
 utomatic parallelization\, data-flow and synchronous programming. He co-au
 thored more than 100 peer-reviewed papers\, is or has been the advisor for
  21 PhD theses and served in the program committees of the major conferenc
 es in the field. Several research projects initiated or led by Albert Cohe
 n resulted in the transfer of advanced compilation techniques to productio
 n compilers.More information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR