IEEE International Conference on

"Introduction to IBM Clustering technologies solutions and methods for High Availability and Disaster Recovery cross system plattforms"

This presentation will provide you an overview of the current status quo on clustering for high availability and disaster recovery designs
in commercial and scientific areas.

9:00	INVITED SPEAKER A Case for using High-level Knowledge for Scalable I/O Optimizations Prof. Alok Choudhary. ECE Department, Northwestern University. Evanston, IL , USA ABSTRACT In this talk we make a case for using high-level knowledge and access patterns to perform I/O optimizations to achieve scalability. In particular, we describe techniques that can reduce and/or eliminate locking by explicitly managing I/O concurrency, which is an important step in achieving true scalability. We present performance results by incorporating these techniques in MPI-IO layer. We also describe high-level caching techniques for scalable I/O.
9:30
10:00	Improving the Performance of Cluster Applications through I/O Proxy Architecture. L. M. Sanchez, A. Calderon, F. Isaila, D. Exposito and J. D. Garcıa
10:15	Efficient Data-Movement for Lightweight I/O. Ron A. Oldfield, Patrick Widener, Arthur B. Maccabe, Lee Ward and Todd Kordenbrock
10:30	Coffee break
11:00	Marching Towards Nirvana: Configurations for Very High Performance Parallel File Systems. Phil Andrews, Chris Jordan, Wayne Pfeiffer.
11:20	Fairness and Performance Isolation: an Analysis of Disk Scheduling Algorithms Seetharami R. Seelam and Patricia J. Teller
11:40	Positioning Dynamic Storage Caches for Transient Data Sudharshan S. Vazhkudai, Douglas Thain, Xiaosong Ma, Vincent W. Freeh
12:00	Using Lightweight Transactions and Snapshots for Fault-Tolerant Services Based on Shared Storage Bricks Michail D. Flouris, Renaud Lachaize, and Angelos Bilas
12:20	Closing

10:00 – 11:30 Heterogeneous computing in P2P systems. Chair Alexey Lastovetsky

10:00-10:10 Alexey Lastovetsky, Presentation of Heteropar06 10:10-10:35 JaceP2P: an Environment for Asynchronous Computations on Peer-to-Peer Networks, Jacques M. Bahi, Raphaël Couturier and Philippe Vuillemin 10:35-11:00 Virtual Structured P2P Network Topology for Distributed Computing, Euloge Edi, Tahar Kechadi and Ronan McNulty

11:30 - 13:10 Models and tools for heterogeneous computing. Chair Antonio J. Plaza

11:30-11:55 Analytical Network Modeling of Heterogeneous Large-Scale Cluster Systems, Bahman Javadi, Jemal H. Abawajy, Saeid Nahavandi and Mohammad K. Akbari 11:55-12:20 A Framework for Adaptive Communication Modeling on Heterogeneous Hierarchical Clusters, Wahid Nasri, Hajer Hamad and Hadhemi Fejjari 12:20-12:45 Open MPI: A High-Performance, Heterogeneous MPI, Richard L. Graham, Galen M. Shipman, Brian W. Barrett, Ralph H. Castain and George Bosilca 12:45-13:10 TGrid – Grid runtime support for hierarchically structured task-parallel programs, Sascha Hunold, Thomas Rauber and Gudula Rünger

14-14:25 A 2-Approximation Algorithm for Scheduling Independent Tasks onto a Uniform Parallel Machine and its Extension to a Computational Grid, Noriyuki Fujimoto and Kenichi Hagihara 14:25-14:50 A Quadratic Self-Scheduling Algorithm for Heterogeneous Distributed Computing Systems, J. Díaz, S. Reyes, A. Niño and C. Muñoz-Caro 14:50-15:15 Self-Adapting Scheduling for Tasks with Dependencies in Stochastic Environments, I. Riakotakis, F. M. Ciorba, T. Andronikos and G. Papakonstantinou

16:00 - 17:40 Algorithms and applications in heterogeneous systems. Chair Serge Petiton

16-16:25 Matrix Multiplication on Two Interconnected Processors, Brett A. Becker and Alexey Lastovetsky 16:25-16:50 A Parallel Algorithm for Solution of the Deconvolution Problem on Heterogeneous Networks, Pedro Alonso, Alexey Lastovetsky and Antonio M. Vidal 16:50-17:15 Well balanced sparse matrix-vector multiplication on a parallel heterogeneous system, Clovis Dongmo Jiogo, Pierre Kuonen and Pierre Manneback 17:15-17:40 Heterogeneous Computing in Remote Sensing Applications: Current Trends and Future Perspectives, Antonio J. Plaza 17:45 - 18:30 Panel: Techniques for the development of efficient parallel routines for heterogeneous systems Tentative panelists: Alexey Lastovetsky, Alexey Kalinov, Ravi Reddy, Serge Petiton, Enrique Quintana

9:00 - 9:45 Dr. Giri.Chukkapalli, SUN Microsystems
“SUN Strategy for Sustained Performance on Multicore Petascale Architectures “

9:45 - 10:30 Dr. Kent Milfeld, Texas Advanced Computing Center
“Multi-Core Programming Concepts in Commodity Systems”

10:30 - 11:00 Coffee Break

11:00 - 11:45 Dr. Herve Chevanne, Advanced Micro Devices
Building Tomorrow’s Clusters with AMD Multi-core Processors

11:45 - 12:30 Dr. Antonio González, Intel
“The Intel Core Microarchitecture and Multi-Core Roadmap”

12:30 - 13:00 Joint Multi-Core Discussio

10:00 - 10:30 "Out of User Space Storage and RDMA", Michael Ko, Renato Recio, and Claudia Salzberg, IBM.

10:30 - 11:00 "Initial Performance Evaluation of the NetEffect 10 Gigabit iWARP Adapter", Dennis Dalessandro, Pete Wyckoff, Ohio Supercomputer Center, and Gary Montry, NetEffect Inc.

11:30 - 12:00 "Exploiting RDMA Operations for Providing Efficient Fine-Grained Resource Monitoring in Cluster-based Servers", K. Vaidyanathan, Ohio State University, H. - W. Jin, Konkuk University, and D. K. Panda, Ohio State University.

12:00 - 12:30 "Software iSCSI over iWARP on a General Purpose CPU", Abhijeet A. Joglekar and Steven King, Intel Corporation.

12:30 - 1:00 "Architecture and Implementation of Sockets Direct Protocol in Windows", Dror Goldenberg, Tzachi Dar, and Gilad Shainer, Mellanox Technologies Inc.

2:00 - 2:30 "Experiences from Debugging a PCIX-based RDMA-capable NIC", Manolis Marazakis, Vassilis Papaefstathiou, Giorgos Kalokairinos, and Angelos Bilas, Institute of Computer Science (ICS).

2:30 - 3:00 "Performance Study of Winsock Direct with 10 Gb Ethernet RDMA-Enabled NIC", Brian Hausauer, NetEffect, Inc.