Call for Papers



Sunday, Febryary 8th, 2015.

 Schedule:


Time

Speaker

Title

8:30-9:00

GT/JF/JB

Introduction

9:00-10:00

Michael Wong, IBM

OpenMP SIMD -  keynote

10:00-10:30


Break

10:30-11:00


James Brodman

Intel


ISPC

11-11:30

David C. Wong, Vincent Palomares, Emmanuel Oseret, Zakaria Bendifallah, Mathieu Tribalat, William Jalby and David J. Kuck


Intel

VP3: A Vectorization Potential Performance Prototype

11:30-12:00

Gabriel Tanase
IBM

ISPC for IBM Power


12:00-14:00


Lunch - sponsored by the conference

14:00-14:30

Jan Treibig


Friedrich-Alexander-Universitaet Erlangen-Nuernberg

Kahan sum with various SIMD instruction sets

14:30-15:00

Peter Jensen, Ningxin Hu, Ivan Jibaja, Dan Gohman and John McCutchan

SIMD in JavaScript via C++ and Emscripten


15:00-15:30

Michael Gschwind and William Schmidt


IBM

Bridging endian-dependent SIMD vector representation with compiler optimization


15:30-16:00



Closing and Break




   





While GPU has gained wide-spread attention in the research community in recent years, the design of “traditional” Single Instruction Multiple Data (SIMD) and vector processing units for modern processors (ARM, POWER, X86) have also quietly gone through an evolution to increase processing power, relax programmability constraints, and expand application domains.  It is time to take a fresh look at the programming interface design of SIMD/vector units, some of which may sound last century like SIMD intrinsics, while others increasingly resemble programming interfaces for GPU.


Scope


The purpose of this workshop is to bring together practitioners and researchers from academia and industry alike to discuss issues, solutions, and opportunities in enabling end users (developers) to effectively exploit SIMD/vector processing in modern processors. We seek submissions that cover all aspects of SIMD/vector processing.  Topics of interests include, but are not restricted to:

  • Programming models for SIMD/vector processing:

    • C/C++/Fortran extensions for SIMD (e.g., OpenMP, OpenACC, OpenCL, SIMD intrinsics)

    • New data parallel or streaming programming models for SIMD

  • Exploitation of SIMD/vector in Java, scripting languages, and domain-specific languages

  • Compilers & tools to discover and optimize SIMD parallelism
  • Application frameworks for developing SIMD/vector applications

  • Case study, experience report, and performance analysis of SIMD/vector applications


Submission


Submitted papers must be no more than 8 pages in length. Authors are encouraged to use the ACM two column format here. Each submission will receive at least three reviews from the technical program committee and authors of selected submissions will have 30 minutes to present their work at the workshop.


Authors must register and submit the paper through online submission system, if you have problems accessing the system, e-mail your submission to igtanase@us.ibm.com .


Important dates


* Submission deadline: 11:59pm EST, December 1

* Extended Submission deadline: 11:59pm EST, December 10



Organizers


Gabriel Tanase (igtanase@us.ibm.com), Joel Falcou (NumScale, Université Paris Sud) and James Brodman (Intel).


Program Committee (In progress)

  • Maria Jesus Garzaran ( UIUC )
  • Vikas Aggarwal ( Oracle Research Labs )
  • Joel Falcou ( NumScale, Université Paris Sud )

  • Gabriel Tanase ( IBM research )

  • James Brodman ( Intel )
  • Sebastian Pop ( Samsung Austin R&D Center )
  • Jan Treibig ( Friedrich-Alexander-Universitaet Erlangen-Nuernber )
  • Jessica H. Tseng ( IBM Research )
  • Sebastian Hack, ( Saarland University )
  • Franz Franchetti ( CMU )
  • Peng Tu (Intel)
  • Chuck Zhao (Samsung Lab)
  • Daniel Etiemble ( Université Paris Sud )
  • Lionel Lacassagne ( Université Paris Sud )