[ecoop-info] [CFP] DEADLINE APPROACHING - JPDC Special Issue on: Architectures and Algorithms for Irregular Applications
Antonino.Tumeo at pnnl.gov
Wed Mar 12 19:15:00 CET 2014
Call for papers
Journal of Parallel and Distributed Computing:
Special Issue on: Architectures and Algorithms for Irregular Applications
There is an emergence of data intensive, irregular applications for knowledge discovery in such diverse fields as cybersecurity, bioinformatics, Computer Aided Design, machine learning, and the semantic analysis of complex social, transportation, and communication networks. Knowledge discovery applications operate on web-scale data sets best represented as graphs using pointer- or linked list based data structures. Consequently, these applications are irregular in both data and control flow. For the most part, they generate concurrent activity per data element and unpredictable, fine-grain communication requests. Since extent supercomputing systems are built with hardware components and software stacks optimized for data locality and regular computation, developing irregular applications for these systems demands a substantial effort and still results in poor execution performance and scalability.
New designs are needed to address irregular application challenges impacting all aspects of the computer hardware and software stack, including micro- and system-architectures, runtime systems, compilers, languages, libraries, and algorithms. Only collaborative efforts among researchers with different expertise can lead to significant breakthroughs. The objective of this special issue is to collect the most novel approaches for irregular applications and to establish a foundation on which future solutions can be built. This special issue will catalyze the surging interest in irregular applications. We will accept both novel unpublished work, as well as published, but significantly extended, work. Articles may address any aspect of the hardware and software stack listed above.
Call for Papers
There is an emergence of data intensive, irregular applications for knowledge discovery in such diverse fields as cybersecurity, bioinformatics, Computer Aided Design, machine learning, and the semantic analysis of complex social, transportation, and communication networks. Knowledge discovery applications operate on web-scale data sets best represented as graphs using pointer- or linked list based data structures. While these applications have a significant degree of latent parallelism, they are difficult to scale on current high-performance computer systems because of their fine-grain, irregular, and unpredictable data accesses. Moreover, their data sets are difficult to partition and generate load imbalances.
Current high performance architectures rely on data locality, regular computations, structured data, and assume datasets that are easy to partitioned. Consequently, they do not support the requirements of irregular applications. Addressing these requirements throughout the hardware and software stack of current and future system architectures will become critical to solving the scientific challenges of the next decade.
This special issue seeks to explore solutions for supporting efficient design, development, and execution of irregular applications in the form of new features for micro- and system- architectures, runtime systems, compilers, languages, libraries, and algorithms. Topics of interest, of both theoretical and practical significance, include but are not limited to:
* Micro- and System-architectures
* Network and memory architectures
* Manycore, hybrid, heterogeneous and custom architectures (Tilera, GPUs, FPGAs) Modeling, evaluation and characterization of architectures for memory intensive and irregular applications
* Innovative algorithmic techniques
* Combinatorial (graph) algorithms and their applications Parallelization techniques and data structures Languages and programming models
* Library and runtime support
* Compiler and analysis techniques
* Case studies of irregular applications (e.g. Semantic Graph Databases, Data Mining, Security, Bioinformatics
This special issue solicits novel, unpublished work, and previously published, but significantly extended, work.
The submitted papers must be written in English and describe original research which is not published nor currently under review by other journals or conferences. Author guidelines for preparation of manuscripts can be found at http://www.elsevier.com/journals/journal-of-parallel-and-distributed-computing/0743-7315/guide-for-authors.
For more information, please contact Antonino Tumeo (antonino.tumeo at pnnl.gov).
All manuscripts and any supplementary material should be submitted through Elsevier Editorial System (EES). The authors must select ‘‘Special Issue: AAIA’’ when they reach the ‘‘Article Type’’ step in the submission process. The EES website is located at:http://ees.elsevier.com/jpdc.
November 15, 2013 to March 24, 2014
July 26, 2014
September 29, 2014
Guide for Authors
This site will guide you stepwise through the creation and uploading of your article. The Guide for Authors can be found on the JPDC journal homepage (http://www.elsevier.com/locate/jpdc).
John T. Feo, PNNL
Antonino Tumeo, PNNL
Timothy G. Mattson, Intel
Oreste Villa, NVIDIA
Simone Secchi, Università di Cagliari
High Performance Computing
Pacific Northwest National Laboratory
antonino.tumeo at pnnl.gov
More information about the ecoop-info