[ecoop-info] Deadline Extended: Workshop on Programming Model and Supportive Environment for Heterogeneous Systems (PMHS)

LIN MA larry.mars at gmail.com
Thu May 28 07:17:49 CEST 2015

*International Workshop on Programming Model and Supportive Environment for
Heterogeneous Systems (PMHS)  *

*in conjunction with The 17th IEEE International Conference on High
Performance Computing and Communications (HPCC 2015)*


*August 24 - 26, 2015, New York, USA*

*Deadline extended to June 10, 2015*

Rapid advancements in heterogeneous architectures have placed incremental
challenges on programming productivity, portability, and performance on
such systems. From high-end servers to mobile devices, from specialized
accelerators to embedded SoCs, from single node multicore/manycore to
distributed clusters in cloud, be it NVIDIA, ARM, Intel, AMD or Cray, there
are a many complex low-level details needed to be managed and abstracted
from the programmers.

The PMHS workshop aims to bring together researchers and industrial
practitioners both from computer architecture scientists and the software
engineering community to present the latest ideas and findings in all
aspects of heterogeneous system programming models, language design or
extension, compilers design or optimization, runtime systems, and
programming tools. The objective is to brainstorm ways to make programming
heterogeneous systems less challenging and more interesting, and promote
the synergies between hardware design and software design.

Topics of interest to the PHMS workshop include but are not limited to:

   - Programming environments for accelerators, embedded/SoC systems, and
   multicore/manycore processors, such as GPUs, FPGAs, Cells, MICs, etc.
   - Programming environments for large-scale parallel systems, clusters of
   multicore/manycore, distributed big-data systems, and SDN.
   - Compiler optimizations and tuning heterogeneous systems
   (Parallelization and loop transformations/fusion, Locality/affinity
   optimizations, SIMDization/Vectorization, Reducing synchronization &
   scheduling overheads, Tiling and offloading)
   - Innovative runtime systems for multicore processors, heterogeneous
   systems, and accelerators.
   - Scalable tools for performance analysis, modeling, monitoring, and
   - Programming language/models: thread and task based models, data
   parallel models, stream programming models, shared/distributed memory
   models, PGAS, etc.
   - Hardware-software co-design,OS and architectural support for parallel
   - Improvements for existing parallel languages and run-time environments
   such as MPI, OpenMP, Cilk, OpenACC, OpenCL, CUDA, ArrayFire, FastFlow,
   OpenSHMEM, UPC, Co-Array Fortran (CAF), Titanium, GASPI, X10, and Chapel,

Manuscript Submission & Proceedings

Papers reporting original and unpublished research results are solicited. *All
submissions will be handled electronically through EasyChair* (
https://easychair.org/conferences/?conf=pmhs2015). Submissions will be
judged by at least 3 reviews on originality, technical strength, quality of
presentation, and relevance to the workshop scope. The length of the
camera-ready manuscripts will be limited to 4 pages (6 pages with the over
length charge) in standard IEEE CS proceedings paper format (two-column).

All accepted papers will be included in the HPCC-2015 Workshop Proceeding
published by IEEE Computer Society (indexed by EI). Authors of selected
papers in PMHS 2015 will be invited to extend their papers for submission
to special issues in prestigious SCI & EI indexed Journals. Authors of
accepted papers, or at least one of them, are required to register and
present their work at the conference, otherwise their papers will not be

*Important Dates:*

*June 10, 2015: Due date for full workshop papers submission(Extended)*

*June 30, 2015:  Notification of paper acceptance to authors*

*July 20, 2015:   Camera-ready of accepted papers*

*August 24-26:   Workshop in conjunction with HPCC 2015*

Workshop Organization


   - Lin Ma Huawei America Research Center, USA lin.ma at huawei.com
   - Rajdeep Bhowmik Cisco, USA  rbhowmik at cs.binghamton.edu

Program Committee (to be extended):

   - Zheng Chen         Facebook, USA
   - Peng Li  Amazon, USA
   - Robert Utterback  Washington University in Saint Louis, USA
   - Chengjie Wu  Yahoo, USA
   - Xingjing Lu ICT, Chinese Academy of Sciences, China
   - Wenlei Bao  Ohio State University, USA
   - Fenglong Song Huawei, China
   - Lingjun Fan ICT, Chinese Academy of Sciences, China
   - Jing Li  Washington University in Saint Louis, USA
   - Jiefan Qiu Zhejiang University, China
   - Shigang Li ICT, Chinese Academy of Sciences, China

More information about the ecoop-info mailing list