Conference Program

Wednesday, August 29

09:30-10:00, Conference Opening 
Room: Elafos

10:00-11:00, Keynote 1
Chair: Christos Kaklamanis // Room: Elafos

Thomas Schulthess, CSCS, ETH Zurich, Switzerland
An applications perspective on multi-core, massive multi-threading, and hybrid systems


11:30-13:00, Session A1 - Parallel and Distributed Programming
Chair: Rizos Sakellariou // Room: Elafos

Assessing the Performance and Scalability of a Novel Multilevel K-nomial Allgather on CORE-Direct Systems
Joshua Ladd, Manjunath Gorentla Venkata, Richard Graham, Pavel Shamis

Hierarchical Partitioning Algorithm for Scientific Computing on Highly Heterogeneous CPU + GPU Clusters
David Clarke, Aleksandar Ilic, Alexey Lastovetsky, Leonel A Sousa

Encapsulating synchronization and load-balance in heterogeneous programming
Yuri Torres, Arturo Gonzalez-Escribano, Diego R. Llanos


11:30-13:00, Session A2 - Performance Prediction and Evaluation
Chair: Helen Karatza // Room: Clio

Energy Consumption Modeling for Hybrid Computing
Ami Marowka

CRAWP: A Workload Partition Method for the Efficient Parallel Simulation of Manycores
Shuai Jiao, Paolo Ienne, Xiaochun Ye, Da Wang, Dongrui Fan, Ninghui Sun

ASK: Adaptive Sampling Kit for Performance Characterization
Pablo de Oliveira Castro, Eric Petit, Jean Christophe Beyler, William Jalby


11:30-13:00, Session A3 - High-Performance Network and Communication
Chair: Kostas Christodoulopoulos // Room: Ourania

An adaptive, scalable, and portable technique for speeding up MPI-based applications
Rosa Filgueira, Malcolm Atkinson, Alberto Nunez, Javier Fernandez

Towards an Efficient Fat-tree like Topology
Diego Bermudez Garzon, Crispin Gomez, Maria Engracia Gomez, Pedro Lopez, Jose Duato

Cost-effective Contention Avoidance in a CMP with Shared Memory Controllers
Samuel Rodrigo, Frank Olaf Sem-Jacobsen, Herve Tatenguem, Tor Skeie, Davide Bertozzi


14:30-16:00, Session B1 - Peer to Peer Computing
Chair: Frederic Desprez // Room: Elafos

Changing the Unchoking Policy for an Enhanced Bittorrent
Vaggelis Atlidakis, Mema Roussopoulos, Alex Delis

Peer-to-Peer Multi-Class Boosting
Istvan Hegedus, Robert Busa-Fekete, Robert Ormandi, Mark Jelasity, Balazs Kegl

ID-Replication for Structured Peer-to-Peer Systems
Tallat M. Shafaat, Bilal Ahmad, Seif Haridi


14:30-16:00, Session B2 - Theory and Algorithms for Parallel Computation - Performance Prediction and Evaluation
Chair: Christos Zaroliagis // Room: Clio

A Lower Bound Technique for Communication on BSP with Application to the FFT
Gianfranco Bilardi, Michele Scquizzato, Francesco Silvestri

A fast parallel algorithm for minimum-cost small integral flows
Andrzej Lingas, Mia Persson

HPC File Systems in Wide Area Networks: Understanding the Performance of Lustre over WAN
Alvaro Aguilera, Michael Kluge, Thomas William, Wolfgang E. Nagel


14:30-16:00, Session B3 - Multicore and Manycore Programming
Chair: Raymond Namyst // Room: Ourania

Task-parallel Programming on NUMA Architectures
Christian Terboven, Dirk Schmidl, Tim Cramer, Dieter an Mey

An Efficient Unbounded Lock-Free Queue for Multi-Core Systems
Marco Aldinucci, Marco Danelutto, Peter Kilpatrick, Massimiliano Meneghin, Massimo Torquati

Speeding Up OpenMP Tasking
Spiros N. Agathos, Nikolaos D. Kallimanis, Vassilios V. Dimakopoulos


16:30-18:00, Session C1 - Mobile and Ubiquitous Computing - Performance Prediction and Evaluation
Chair: Sotiris Nikoletseas // Room: Elafos

Watershed-based Clustering for Energy Efficient Data Gathering in Wireless Sensor Networks with Mobile Collector
Charalampos Konstantopoulos, Basilis Mamalis, Grammati Pantziou, Vasileios Thanasias

Distribution of Liveness Property Connectivity Interval in Selected Mobility Models of Wireless Ad Hoc Networks
Michal Kalewski, Jerzy Brzezinski, Marek Libuda, Marcin Kosiba

Understanding I/O Performance using I/O Skeletal Applications
Jeremy Logan, Scott Klasky, Hasan Abbasi, Qing Liu, George Ostrouchov, Manish Parashar, Norbert Podhorszki, Yuan Tian, Matthew Wolf


16:30-18:00, Session C2 - Scheduling and Load Balancing
Chair: Chryssis Georgiou // Room: Clio

Job Scheduling Using successive Linear Programming Approximations of a Sparse Model
Stephane Chretien, Jean-Marc Nicod, Laurent Philippe, Veronika Rehn-Sonigo, Lamiel Toch

Speed Scaling on Parallel Processors with Migration
Eric Angel, Evripidis Bampis, Fadi Kacem, Dimitrios Letsios

Dynamic Distributed Scheduling Algorithm for State Space Search
Ankur Narang, Abhinav Srivastava, Ramnik Jain, R. K. Shyamasundar


16:30-18:00, Session C3 - Multicore and Manycore Programming
Chair: Emmanuel Jeannot // Room: Ourania

High-Level Support for Pipeline Parallelism on Manycore Architectures
Siegfried Benkner, Enes Bajrovic, Erich Marth, Martin Sandrieser, Raymond Namyst, Samuel Thibault

Node.Scala: Implicit Parallel Programming for High-Performance Web Services
Daniele Bonetta, Danilo Ansaloni, Achille Peternier, Cesare Pautasso, Walter Binder

Folding of Tagged Single Assignment Values for Memory-Efficient Parallelism
Dragos Sbirlea, Kath Knobe, Vivek Sarkar



Thursday, August 30


9:30-10:30, Keynote 2
Chair: Theodore Papatheodorou // Room: Elafos

Ewa Deelman, Information Sciences Institute, USC, USA
Hosted Science: Managing Computational Workflows in the Cloud


11:00-13:00, Session D1 - Scheduling and Load Balancing
Chair: Alexey Lastovetsky // Room: Elafos

Using Load Information in Work-Stealing on Distributed Systems with Non-Uniform Communication Latencies
Vladimir Janjic, Kevin Hammond

Energy Efficient Frequency Scaling and Scheduling for Malleable Tasks
Jochen Speck, Peter Sanders

Scheduling MapReduce jobs in HPC clusters
Marcelo Neves, Tiago Ferreto, Cesar A. F. De Rose

A job scheduling approach for multi-core clusters based on virtual malleability
Gladys Utrera, Siham Tabik, Julita Corbalan, Jesus Labarta



11:00-13:00, Session D2 - Support Tools and Environments

Chair: Marios Dikaiakos // Room: Aphroditi

Tulipse: A Visualization Framework for User-Guided Parallelization
Yi Wen Wong, Wai Teng Tang, Wen Jun Tan, Tomasz Dubrownik, Rubing Duan, Rick Siow Mong Goh, Shyh-hao Kuo, Stephen John Turner, Weng Fai Wong

Enabling cloud interoperability with COMPSs
Fabrizio Marozzo, Francesc J Lordan, Roger Rafanell, Daniele Lezzi, Domenico Talia, Rosa M. Badia

Pattern-Independent Detection of Manual Collectives in MPI Programs
Alexandru Calotoiu, Christian Siebert, Felix Wolf

A Type-based Approach to Separating Protocol from Application Logic: A Case Study in Hybrid Computer Programming
Geoffrey Hulette, Allen D. Malony, Matthew Sottile



11:00-13:00, Session D3 - Parallel and Distributed Data Management

Chair: Domenico Talia // Room: Clio

3D Inverted Index with Cache Sharing for Web Search Engines
Esteban Feuerstein, Graciela Veronica Gil-Costa, Mauricio Marin, Gabriel Tolosa, Riccardo Baeza-Yates

Quality-of-Service for Consistency of Data Geo-Replication in Cloud Computing
Sergio Esteves, Joao N Silva, Luis Veiga

DS-Means: Distributed Data Stream Clustering
Alessio Guerrieri, Alberto Montresor

A Fault-Tolerant Cache Service for Web Search Engines
Carlos Gomez-Pantoja, Dolores Rexachs, Mauricio Marin, Emilio Luque



11:00-13:00, Session D4 - High-Performance Architecture and Compilers

Chair: Nectarios Koziris // Room: Ourania

Exploiting Semantics of Virtual Memory to Improve the Efficiency of the On-Chip Memory System
Bin Li, Zhen Fang, Li Zhao, Xiaowei Jiang, Lin Li, Andrew Herdrich, Ravishankar Iyer, Srihari Makineni

Dynamic Last-Level Cache Allocation to Reduce Area and Power Overhead in Directory Coherence Protocols
Mario Lodde, Jose Flich, Manuel E. Acacio

From Serial Loops to Parallel Execution on Distributed Systems
George Bosilca, Aurelien Bouteiller, Anthony Danalis, Thomas Herault, Jack Dongarra

A Practical Approach to DOACROSS Parallelization
Priya Unnikrishnan, Jun Shirako, Kit Barton, Sanjay Chatterjee, Raul Silvera, Vivek Sarkar



14:30-16:00, Session E1 - High Performance and Scientific Applications

Chair: Thomas Ludwig // Room: Elafos

Memory-Access Optimization of Parallel Molecular Dynamics Simulation via Dynamic Data Reordering
Manaschai Kunaseth, Ken-ichi Nomura, Hikmet Dursun, Rajiv Kalia, Aiichiro Nakano, Priya Vashishta

On Analyzing Quality of Data Influences on Performance of Finite Elements driven Computational Simulations
Michael Reiter, Hong-Linh Truong, Schahram Dustdar, Dimka Karastoyanova, Robert Krause, Frank Leymann, Dieter Pahr

Performance Evaluation and Optimization of Nested High Resolution Weather Simulations
Preeti Malakar, Vaibhav Saxena, Thomas George, Rashmi Mittal, Sameer Kumar, Abdul Naim, Saiful A. Husain



14:30-16:00, Session E2 - Parallel and Distributed Programming

Chair: Marco Danelutto // Room: Aphroditi

Transactional access to shared memory in StarSs, a task based programming model
Rahulkumar Gayatri, Rosa M. Badia, Eduard Ayguade, Mikel Lujan, Ian Watson

On-the-fly task execution for speeding up pipelined MapReduce
Diana Moise, Gabriel Antoniu, Luc Bouge

Dynamic Thread Mapping Based on Machine Learning for Transactional Memory Applications
Marcio Castro, Luis Fabricio Wanderley Goes, Luiz G. Fernandes, Jean-Francois Mehaut



14:30-16:00, Session E3 - Distributed Systems and Algorithms

Chair: Ioannis Caragiannis // Room: Clio

From a Store-collect Object and $\Omega$ to Efficient Asynchronous Consensus
Michel Raynal, Julien Stainer

An Investigation into the performance of reduction algorithms under load imbalance
Petar Marendic, Jan Lemeire, Tom Haber, Dean Vucinic, Peter Schelkens

Achieving Reliability in Master-worker Computing via Evolutionary Dynamics
Evgenia Christoforou; Antonio Fernandez Anta; Chryssis Georgiou; Miguel A. Mosteiro; Angel (Anxo) Sanchez



14:30-16:00, Session E4 - GPU and Accelerators Computing

Chair: Bernd Mohr // Room: Ourania

A new programming paradigm for GPGPU
Julio Toss, Thierry Gautier

GPURoofline: A Model for Guiding Performance Optimizations on GPUs
Haipeng Jia, Yunquan Zhang, Long Guoping, Jianliang Xu, Shengen Yan, Yan Li

Understanding the Performance of Concurrent Data Structures on Graphics Processors
Daniel Cederman, Bapi Chatterjee, Philippas Tsigas



16:30-18:00, Session Distinguished Papers

Chair: Christian Lengauer // Room: Elafos

Topology Configuration in Hybrid EPS/OCS Interconnects
Kostas Christodoulopoulos, Marco Ruffini, Donal O'Mahony, Kostas Michael Katrinis

A Checkpoint-on-Failure Protocol for Algorithm-Based Recovery in Standard MPI
Wesley Bland, Peng Du, Aurelien Bouteiller, Thomas Herault, George Bosilca, Jack Dongarra

Efficient Support for In-Place Metadata in Transactional Memory
Ricardo J. Dias, Joao M. S. Lourenco, Tiago Vale



Friday, August 31


09:30-11:00, Session F1 - High Performance and Scientific Applications
Chair: Alice Koniges // Room: Elafos

GPU-vote: A Framework for Accelerating Voting Algorithms on GPU
Gert-Jan van den Braak, Cedric Nugteren, Bart Mesman, Henk Corporaal

Topology-aware Mappings for Large-Scale Eigenvalue Problems
Hasan Metin Aktulga, Chao Yang, Esmond G Ng, Pieter Maris, James Vary

Fast and Effective Lossy Compression Algorithms for Scientific Datasets
Jeremy Iverson, Chandrika Kamath, George Karypis



09:30-11:00, Session F2 - GPU and Accelerators Computing

Chair: Nectarios Koziris // Room: Clio

GPU-Accelerated Asynchronous Error Correction for Mixed Precision Iterative Refinement
Hartwig Anzt, Piotr Luszczek, Jack Dongarra, Vincent Heuveline

OpenACC - First Experiences with Real-World Applications
Sandra Wienke, Paul Springer, Christian Terboven, Dieter an Mey

Building a Collision for 75-Round Reduced SHA-1 Using GPU Clusters
Andrew Adinetz, Evgeny Grechnikov



09:30-11:00, Session F3 - Distributed Systems and Algorithms - Grid, Cluster and Cloud Computing

Chair: Marios Mavronicolas // Room: Ourania

Towards Load Balanced Distributed Transactional Memory
Gokarna Sharma, Costas Busch

CUDA-For-Clusters: A System for Efficient Execution of CUDA Kernels on Multi-Core Clusters
Raghu Prabhakar, Govindarajan Ramaswamy, Matthew Jacob Thazhuthaveetil

Caching VM Instances for Fast VM Provisioning: A Comparative Evaluation
Pradipta De, Manish Gupta, Manoj Soni, Aditya Thatte



11:30-13:00, Session G1 - Parallel Numerical Algorithms

Chair: Efstratios Gallopoulos // Room: Elafos

Locality Improvement of Data-Parallel Adams-Bashforth Methods Through Block-Based Pipelining of Time Steps
Matthias Korch

Avoiding communication through a multilevel LU factorization
Simplice Donfack, Laura Grigori, Amal Khabou

Parallel SOR for solving the Convection Diffusion equation using GPUs with CUDA
Yiannis Cotronis, Elias Konstantinidis, Maria Louka, Nikolaos Missirlis



11:30-12:30, Session G2 - GPU and Accelerators Computing

Chair: Peter Kilpatrick // Room: Clio

Optimized Hybrid Parallel Lattice Boltzmann Fluid Flow Simulations on Complex Geometries
Jonas Fietz, Mathias Krause, Christian Schulz, Peter Sanders, Vincent Heuveline

accull: An OpenACC implementation with CUDA and OpenCL support
Ruyman Reyes, Ivan Lopez-Rodriguez, Juan Fumero, Francisco de Sande



11:30-13:00, Session G3 - Grid, Cluster and Cloud Computing

Chair: Kostas Christodoulopoulos // Room: Ourania

Improving Scheduling Performance using a Q-Learning-based Leasing Policy for Clouds
Alexander Folling, Matthias Hofmann

Impact of Variable Priced Cloud Resources on Scientific Workflow Scheduling
Simon Ostermann, Radu Prodan

Scalable Reed-Solomon-based Reliable Local Storage for HPC Applications on IaaS Clouds
Leonardo Bautista Gomez, Bogdan Nicolae, Naoya Maruyama, Satoshi Matsuoka, Franck Cappello


14:30-15:30, Keynote 3
Chair: Paul Spirakis // Room: Elafos

Burkhard Monien, University of Paderborn, Germany
Selfish distributed optimization



15:30-16:00, Conference Closing

Room: Elafos

cti