The 3rd MATH-EX Workshop in the series is part of the International conference in Computational Science (ICCS2016), to be held in San Diego, California, USA, 6-8 June 2016 (http://www.iccs-meeting.org/iccs2016/). The theme for ICCS 2016 is “Data through the Computational Lens”. ICCS is an ERA 2010 A-ranked conference series.
The workshop aims to serve as a forum for Computational Scientists to discuss the mathematical and algorithmic challenges and approaches towards increasingly larger systems.
Key science applications using HPC systems require novel mathematics and mathematical modelling approaches together with scalable scientific algorithms. This is especially true as HPC systems continue to scale up in compute node and processor core count. These novel mathematical methods can lead to scalable scientific algorithms which can hide network and memory latency, have very high computation/communication overlap and minimal communication and fewer synchronization points.
The workshop solicits strategic and position papers presenting Domain Applications and Case Studies in the context of the Computational Science/ HPC ecosystem covering but not limited to the following topics:
The workshop aims to serve as a forum for Computational Scientists to discuss the mathematical and algorithmic challenges and approaches towards increasingly larger systems.
Key science applications using HPC systems require novel mathematics and mathematical modelling approaches together with scalable scientific algorithms. This is especially true as HPC systems continue to scale up in compute node and processor core count. These novel mathematical methods can lead to scalable scientific algorithms which can hide network and memory latency, have very high computation/communication overlap and minimal communication and fewer synchronization points.
The workshop solicits strategic and position papers presenting Domain Applications and Case Studies in the context of the Computational Science/ HPC ecosystem covering but not limited to the following topics:
- Novel scientific algorithms that improve performance, scalability, resilience, and power efficiency
- Naturally fault tolerant, self-healing, or fault oblivious scientific algorithms
- Scientific algorithms that can exploit extreme concurrency
- Novel Mathematical methods and mathematical modelling approaches that possess scalability properties:
- Mathematical and algorithmic approaches to resolve the challenges of increased scale of HPC systems.
- Crosscutting approaches, e.g. mathematical methods and algorithmic approaches addressing scalability challenges
Important dates:
- Paper submission deadline: February 10th, 2016
- Notification of acceptance: February 28, 2016
Program Committee:
- Vassil Alexandrov, ICREA-BSC, Spain
- Jack Dongarra, University of Tennessee/ORNL, USA
- Al Geist, ORNL, USA
- Mike Heroux, Sandia National Lab, USA
- Nahid Ernad, Universite de Versailles-Saint-Quentin, France
- Svetozar Margenov, IICT - BAS, Bulgaria
How to Submit:
- Prepare the manuscripts in Procedia Computer Science format (no more than 10 pages)
- Log-in to ICCS2015 Submission in ICCS15 https
- You will need to create an account if you do not have an existing account in EasyChair.
- Specify the workshop name as Mathematical Methods and Algorithms for Extreme Scale
Authors are invited to submit manuscripts describing original, unpublished research and recent developments as well as position and strategic papers in the remit of the Workshop.
All accepted papers will be printed in the conference proceedings published by Elsevier Science in the open-access Procedia Computer Science and indexed by Scopus, ScienceDirect, Thomson Reuters Conference Proceedings Citation (former ISI Proceedings) -an integrated index within Web of Science. The papers will contain linked references, XML versions and citable DOI numbers.
The submitted paper must be camera-ready and formatted according to the rules of Procedia Computer Science, not exceeding 10 pages. Submission implies the willingness of at least one of the authors to register and present the paper.
Papers must be based on unpublished original work and must be submitted to ICCS/its workshops only.
After the conference, selected papers will be invited for a special issue of the Journal of Computational Science.