| Time | Event |
|---|---|
| 9:00-9:05 | Welcome and Introduction Tumeo, Feo, Becchi, Verbanescu |
| 9:05 ~ 9:50 | Invited Talk 1 Chair: Michela Becchi (NCSU) Accelerating Irregular Algorithms on GPUs Martin Burtscher (Texas State University) |
| 9:50~10:00 | Session 1: Sparse Algorithms on Heterogeneous Systems Chair: Ankur Limaye (PNNL) Batch Updates of Distributed Streaming Graphs using Linear Algebra (short presentation) Hassani, Hussain, Azad |
| 10:00~10:30 | Coffee Break |
| 10:30~11:20 | Session 1: Sparse Algorithms on Heterogeneous Systems Chair: Ankur Limaye (PNNL) Establish the basis for Breadth-First Search on Frontier System: XBFS on AMD GPUs Yang, Lu, Sattar, Liu, Wang xBS-GNN: Accelerating Billion-Scale GNN Training on FPGA Lin, Xu, Prasanna |
| 11:20~12:30 | Session 2: Optimizing Sparse Algorithms for Parallel Architectures Chair: Nicolas Bohm Agostini (PNNL) Enhancing Scalability and Performance in Influence Maximization with Optimized Parallel Processing Wu, Xu, Park, Jahan Tithi, Checconi, Wolfson-Pou, Petrini, Krishna Enhancing Small Message Aggregation with Directive-Based Deferred Execution (short paper) Welch, Hernandez, Poole, Poole Shared Memory-Aware Latency-Sensitive Message Aggregation for Fine-Grained Communication (short paper) Chandrasekar, Kale NEO-DNND: Communication-Optimized Distributed Nearest Neighbor Graph Construction Iwabuchi, Steil, Priest, Pearce, Sanders |
| 12:30~14:00 | Lunch Break |
| 14:00~14:15 | Graph500 List Announcement Rich Murphy, Anton Korz |
| 14:15~14:55 | Invited Talk 2 Chair: Vito Giovanni Castellana (PNNL) Discussion on Hash-Table Approaches for Efficient Sparse Tensor Contraction Jiajia Li (North Carolina State University) |
| 14:55~15:15 | Session 3: Programming models for parallel algorithms Chair: Ian Di Dio Lavore (Politecnico di Milano) An Adaptive Asynchronous Approach for the Single-Source Shortest Paths Problem (short paper) Rao, Chandrasekar, Kale GPU Accelerated Sparse Cholesky Factorization (short paper) Karsavuran, Ng, Peyton |
| 15:15~15:30 | Coffee Break |
| 15:30~16:20 | Session 3: Programming models for parallel algorithms (continued) Chair: Ian di Dio Lavore (Politecnico di Milano) Efficient Tree-based Parallel Algorithms for N-Body Simulations Using C++ Standard Parallelism Lane Cassell, Deakin, Alpay, Heuveline, Brito Gadeschi Linear Algebra Approach for Directed Triad Counting and Enumeration Kerdcharoen, Sridhar, Sangpetch, Low |
| 16:20~16:50 | Session 4: Evaluating performance and availability of parallel architectures Chair: Gabriel Rodriguez Canal (University of Edinburgh and PNNL) Performance evaluation and modelling of single-precision matrix multiplication on Cerebras CS-2 (short paper) Matsuzaki, Mukunoki, Miyajima Performance Analysis of the NICAM Benchmark on MN-Core Processor (short paper) Takayashiki, Saito, Imachi, Sakamoto, Makino Predicting Compute Node Unavailability in HPC: A Graph-Based Machine Learning Approach (short paper) Rožanec |
| 16:50-17:30 | IA^3 Debate Moderator: Antonino Tumeo (PNNL) Panelists: Ariful Azad (Indiana University), Giulia Guidi (Cornell University), John Leidel (Tactical Computing Laboratories), Georgios Michelogiannakis (LBNL) |
| 17:30 | Concluding Remarks Tumeo, Feo, Becchi, Verbanescu |
AI Website Generator