<< Go Home

[Click here to download the Program in pdf]

PDSEC-15 Advance Program

Friday, May 29, 2015

Hyderabad International Convention Centre, Hyderabad, INDIA

08:30 - 08:35 am

Welcome and Opening Remarks
Chair: Peter Strazdins

08:35 - 09:30

Keynote talk

Domain-Specific Approaches in Scientific Computing
Naoya Maruyama

09:30 - 10:00

Session 1: Best Paper
Chair: Peter Strazdins


Energy Consumption Reduction with DVFS for Message Passing Iterative Applications on Heterogeneous Architectures
Jean-claude Charr; Raphael Couturier*; Ahmed Fanfakh; Arnaud Giersch

10:00 - 10:30

Tea Break

10:30 - 12:00

Session 2: Performance
Chair: Takeshi Fukaya

Quantifying the Effects of Contention on Parallel File Systems
Steven A. Wright* and Stephen A. Jarvis

Highly Scalable Algorithms for the Sparse Grid Combination Technique
Peter E Strazdins*; Md Mohsin Ali; Brendan Harding

Predicting Optimal Power Allocation for CPU and DRAM Domains
Ananta N Tiwari*; Martin Schulz; Laura Carrington

12:00 - 13:00

Lunch break

13:00 - 15:00

Session 3: Linear Algebra
Chair: Steven Wright


Performance evaluation of the EigenExa eigensolver on Oakleaf-FX: tridiagonalization versus pentadiagonalization
Takeshi Fukaya*; Toshiyuki Imamura


A Resilient Framework for Iterative Linear Algebra Applications in X10
Sara S Hamouda*; Josh Milthorpe; Peter E Strazdins; Vijay Saraswat


Combining backward and forward recovery to cope with silent errors in iterative solvers
Massimiliano Fasi; Yves Robert; Bora Uçar*


TSIRM: A Two-Stage Iteration with least-squares Residual Minimization algorithm to solve large sparse linear systems
Raphael Couturier*; Lilia Ziane Khodja; Christophe Guyeux

15:00 - 15:30

Tea Break

15:30 - 17:00





Session 4: GPUs and Manycore
Chair: Raphael Couturier


Modeling Cooperative Threads to Project GPU Performance for Adaptive Parallelism
Jiayuan Meng; Thomas Uram; Vitali Morozov*; Venkatram Vishwanath; Kalyan Kumaran


GPU Accelerated Molecular Dynamics with Method of Heterogeneous Load Balancing
Takuro Udagawa*; Masakazu Sekijima


Parallel Methods for Optimizing High Order Constellations on GPUs
Paolo Spallaccini*; Farbod Kayhan; Stefano Chinnici; Guido Montorsi


17:00 - 17.05

Closing Remarks
Chair: Raphael Couturier