Program
| Time |
Event |
(+)
|
|
12:30 - 14:00
|
Lunch |
|
|
14:45 - 15:00
|
Introduction - G. Pichon & L. Marchal |
|
|
15:00 - 15:30
|
Programming Heterogeneous Architectures Using Hierarchical Tasks - Gwénolé Lucas |
|
|
15:30 - 16:00
|
Mixed Precision LU Factorization using GPU Tensor Cores - Florent Lopez |
|
|
16:00 - 16:30
|
Task-Based Parallel Programming for Scalable Algorithms - Antoine Jégo |
|
|
16:30 - 17:00
|
Refreshment |
|
|
17:00 - 17:30
|
Learning Optimal Schedules in Markovian Bandits: Pitfalls and Benefits. - Bruno Gaujal |
|
|
17:30 - 18:00
|
Reservation strategies for stochastic jobs - Guillaume Pallez |
|
|
18:00 - 18:30
|
Multi-Phase Task-Based HPC Applications: Quickly Learning how to Run Fast - Arnaud Legrand |
|
|
19:30 - 21:00
|
Dinner |
|
| Time |
Event |
(+)
|
|
09:00 - 09:30
|
Scheduling Strategies for Overloaded Real-Time Systems. - Frédéric Vivien |
|
|
09:30 - 10:00
|
Algorithm-Based Fault Tolerance for Parallel Stencil Computations - Aurélien Cavelan |
|
|
10:00 - 10:30
|
Multi-objective Hybrid Autoscaling of Microservices in Kubernetes Clusters - Hamid Fard |
|
|
10:30 - 11:00
|
Coffee break |
|
|
11:00 - 11:30
|
Mapping Tree-shaped Workflows on Memory-heterogeneous Architectures - Svetlana Kulagina |
|
|
11:30 - 12:00
|
Process Mapping on any Topology with TopoMatch - Emmanuel Jeannot |
|
|
12:30 - 14:00
|
Lunch |
|
|
15:30 - 16:00
|
Data Distribution for Symmetric Linear Algebra Kernels - Mathieu Vérité |
|
|
16:00 - 16:30
|
IO-sets: Simple and efficient approaches for I/O bandwidth management - Nicolas Vidal |
|
|
16:30 - 17:00
|
Memory Aware Scheduling of Tasks Sharing Data on Multiple GPUs with Dynamic Runtime Systems - Maxime Gonthier |
|
|
17:00 - 17:30
|
Refreshment |
|
|
17:30 - 18:00
|
A Selective Nesting Approach for the Sparse Multi-threaded Cholesky Factorization - Valentin Le Fèvre |
|
|
18:00 - 18:30
|
Combining Uncore Frequency and Dynamic Power Capping to Improve Power Savings - Amina Guermouche |
|
|
18:30 - 19:00
|
On the use of simulation for scheduling (research) - Henri Casanova |
|
|
19:30 - 21:00
|
Dinner |
|
| Time |
Event |
(+)
|
|
09:00 - 09:30
|
Investigations of Low-Rank Clustering Techniques for Sparse Direct Solvers - Brian Bantsoukissa |
|
|
09:30 - 10:00
|
Modeling Memory Contention between Communications and Computations in Distributed HPC Systems - Philippe Swartvagher |
|
|
10:00 - 10:30
|
Designing Next-Generation Linear Algebra Libraries for Large-Scale Matrix Computations - Hatem Ltaief |
|
|
10:30 - 11:00
|
Coffee break |
|
|
11:00 - 11:30
|
Memory Saving Techniques for Deep Neural Network Training - Lionel Eyraud-Dubois |
|
|
11:30 - 12:00
|
An Efficient Parallel Implementation of a Perfect Hashing Method for Hypergraphs - Somesh Singh |
|
|
12:30 - 14:00
|
Lunch |
|
|
14:00 - 17:00
|
Excursion |
|
|
19:30 - 21:00
|
Dinner |
|
| Time |
Event |
(+)
|
|
09:00 - 09:30
|
An Exact Algorithm for the Linear Tape Scheduling Problem - Valentin Honoré |
|
|
09:30 - 10:00
|
Shelf schedules for independent moldable tasks to minimize the energy consumption - Anne Benoit |
|
|
10:00 - 10:30
|
Fixed-Parameter Tractability of Scheduling Dependent Tasks on m identical processors subject to Release Times and Deadlines - Alix Munier-Kordon |
|
|
10:30 - 11:00
|
Coffee break |
|
|
11:00 - 11:30
|
Learning-Augmented Online Algorithms - Bertrand Simon |
|
|
11:30 - 12:00
|
Resilient Scheduling of Moldable Parallel Jobs to Cope with Silent Errors - Lucas Pérotin |
|
|
12:30 - 14:00
|
Lunch |
|
|
