| ICS06 Program | ||
| Tuesday | ||
| 16:00 | Registration | Cairns International Hotel |
| 18:00 | Reception | Cairns International Hotel |
| Wednesday | ||
| 8:00-8:30 | Registration | |
| 8:30-8:40 | Welcome | |
| 8:40-9:40 | KeyNote 1 | A Modern High-Performance Processor Pipeline |
| 9:40-9:50 | Coffee Break | |
| 9:50-11:05 | Session 1 | Checkpointing & Speculation |
| Experimental Evaluation of Application-level Checkpointing for OpenMP Programs | ||
| Cooperative Checkpointing: A Robust Approach to Large-Scale Systems Reliability | ||
| On the Performance Potential of Different Types of Speculative Thread-Level Parallelism | ||
| 11:05-11:15 | Coffee Break | |
| 11:15-12:30 | Session 2 | Prediction |
| BranchTap: Improving Performance with Very Few Checkpoints Through Adaptive Speculation Control | ||
| Selective Predicate Prediction for Out-of-Order Processors | ||
| Wide and Efficient Trace Prediction using the Local Trace Predictor | ||
| 12:30-13:45 | LUNCH | |
| 13:45-15:00 | Session 3 | Benchmarking & Modeling |
| Scientific Applications vs. SPEC-FP: A Comparison of Program Behavior | ||
| The Exigency of Benchmark and Compiler Drift: Designing TomorrowÕs Processors with YesterdayÕs Tools | ||
| Accurate Memory Data Flow Modeling in Statistical Simulation | ||
| 15:00-15:10 | Coffee Break | |
| 15:10-16:25 | Session 4 | I/O - Communication |
| Efficient Remote Block-level I/O over an RDMA-capable NIC | ||
| A Scalable Communication Layer for Multi-Dimensional Hyper Crossbar Network Using Multiple Gigabit Ethernet | ||
| Large Files, Small Writes, and pNFS | ||
| 16:25-16:35 | Coffee Break | |
| 16:35-17:50 | Session 5 | High Performance Computing - Supercomputing |
| A Case for High Performance Computing with Virtual Machines | ||
| Implementing Virtual Memory in a Vector Processor with Software Restart Markers | ||
| Multigrid and Gauss-Seidel Smoothers Revisited: Parallelization on Chip Multiprocessors | ||
| Thursday | ||
| 8:30-9:30 | KeyNote 2 | Quantum Mechanical Approaches to Information Processing |
| 9:30-10:45 | Session 6 | Power-Performance |
| Online Power-Performance Adaptation of Multithreaded Programs using Hardware Event-Based Prediction | ||
| A Scalable Low Power Issue Queue for Large Instruction Window Processors | ||
| Design Space Exploration for Multicore Architectures: A Power/Performance/Thermal View | ||
| 10:45-10:55 | Coffee Break | |
| 10:55-12:35 | Session 7 | Multicore Interconnection/Communication |
| Design Tradeoffs for Tiled CMP On-Chip Networks | ||
| STAR-MPI: Self Tuned Adaptive Routines for MPI Collective Operations | ||
| Scaling MPI to short-memory MPPs such as BG/L | ||
| Scalable, Fault Tolerant Membership for MPI Tasks on HPC Systems | ||
| 12:35 | Green Island Great Barrier Reef (Boarding commences 12:30 - 10 minutes walk to harbour) | |
| 19:00 | Banquet at Tjapukai | |
| Friday | ||
| 8:30-9:45 | Session 8 | Memory I |
| Coupling Prefix Caching and Collective Downloads for Remote Dataset Access | ||
| Heterogeneous Way-Size Cache | ||
| Profitable Loop Fusion and Tiling Using Model-driven Empirical Search | ||
| 9:45-10:00 | Coffee Break | |
| 10:00-11:15 | Session 9 | Memory II |
| TMA: A Trap-Based Memory Architecture | ||
| Scalable Algorithms for Global Snapshots in Distributed Systems | ||
| Feedback-directed Memory Disambiguation Through Store Distance Analysis | ||
| 11:15-11:25 | Coffee Break | |
| 11:25-12:40 | Session 10 | Applications |
| Accelerator Design for Protein Sequence HMM Search | ||
| A Distributed System Based on Web Services for Computational Science Simulations | ||
| Accelerating Sparse Matrix Computations via Data Compression | ||
| 12:40-14:00 | LUNCH | |
| 14:00-15:15 | Session 11 | Miscellaneous |
| Sensitivity Analysis of Knapsack-based Task Scheduling on the Grid | ||
| Probabilistic Accuracy Bounds for Fault-Tolerant Computations that Discard Tasks | ||
| Violated Dependence Amalysis | ||
| 15:15-15:30 | Coffee Break | |
| 15:30-16:45 | Session 12 | Scheduling & Mapping |
| User-Guided Symbiotic Space-Sharing of Real Workloads | ||
| MPIPP: An Automatic Profile-guided Parallel Process Placement Toolset in SMP Clusters and Multiclusters | ||
| Lightweight Lock-Free Synchronization Methods for Multithreading | ||
| 16:45-17:00 | Conf. Close | |
| Saturday | ||
| Workhsops and Tutorials | ||