Slides

Recent presentations

• LIP seminar, Lyon, France, May 2023. Revisiting checkpointing techniques and I/O bandwidth-sharing strategies for HPC platforms [pdf].
• Scheduling Variable Capacity Resources for Sustainability, Paris Center of the University of Chicago, France, March 2023. Scheduling algorithms for variable capacity resources [pdf].
• Exposé pour la semaine sport-études des étudiants du L3IF de l'ENS Lyon, Janvier 2023. Comment trouver un sujet de recherche? Replication is more efficient than you think [pdf].
• Keynote talk at the IEEE 34th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), Bordeaux, France, November 2022. Handling Failures on High Performance Computing Platforms: Checkpointing and Scheduling Techniques [pdf].
• Clusters, Clouds, and Data for Scientific Computing, La Maison des Contes, France, September 2022. Resilient scheduling of parallel jobs [pdf].
• 15th Scheduling for Large-Scale Systems Workshop, Fréjus, France, June 2022. Scheduling independent moldable tasks to minimize the energy consumption [pdf].
• Keynote talk at the 18th International Conference on Network and Parallel Computing (IFIP NPC), Paris, November 2021. Resilient scheduling for high-performance computing [pdf].
• FONDA Lecture Series (Foundations of Workflows for Large-Scale Scientific Data Analysis), Online, June 2021. Resilient scheduling on failure-prone platforms [pdf].
• Exposé pour la semaine sport-études des étudiants du L3IF de l'ENS Lyon, Mai 2021. Comment trouver un sujet de recherche? Replication is more efficient than you think [pdf].
• 35th IEEE International Parallel and Distributed Processing Symposium (IPDPS) 2021, Virtual, May 2021. Max-stretch minimization on an edge-cloud platform [pdf-short-version | pdf-long-version].
• Café gourmand et scientifique du LIP, Lyon, Décembre 2020 (online). Quand des erreurs se produisent dans les supercalculateurs... [pdf].
• LSV Seminar, Paris-Saclay, December 2020 (online). Resilient scheduling on failure-prone platforms [pdf].

Slides from 2007-2019

• Keynote talk at the 13th International Conference on Parallel Processing and Applied Mathematics (PPAM), Bialystok, Poland, September 2019. On the scheduling of graphs of tasks: A scalable clustering-based approach using DAG partitioning [pdf].
• 13th International Conference on Parallel Processing and Applied Mathematics (PPAM), Bialystok, Poland, September 2019. Improving locality-aware scheduling with acyclic directed graph partitioning [pdf].
• 33rd IEEE International Parallel and Distributed Processing Symposium (IPDPS) 2019, Rio de Janeiro, Brazil, May 2019. A scalable clustering-based task scheduler for homogeneous processors using DAG partitioning [pdf].
• Panel at HCW workshop, at IPDPS'19, Rio de Janeiro, Brazil, May 2019. Heterogeneity in computing: Now and in the future [pdf].
• Keynote talk at PDCO workshop, at IPDPS'19, Rio de Janeiro, Brazil, May 2019. Optimization problems in the presence of failures on large-scale parallel systems [pdf].
• Keynote talk at 4th GDR RSD and ASF Winter School on Distributed Systems and Networks, Le Pleynet, France, February 2019. Resilient and energy-aware scheduling algorithms [pdf].
• Clusters, Clouds, and Data for Scientific Computing, La Maison des Contes, France, September 2018. Co-scheduling HPC workloads on cache-partitioned CMP platforms [pdf].
• ICL lunch talk, UT Knoxville, TN, USA, June 2018. Combining checkpointing and replication for reliable execution of linear workflows [pdf].
• Diversity panel, at IPDPS'18, Vancouver, Canada, May 2018. My three slides [pdf].
• APDCM Workshop, at IPDPS'18, Vancouver, Canada, May 2018. Combining checkpointing and replication for reliable execution of linear workflows [pdf].
• Invited talk at the HPPAC workshop, at IPDPS'18, Vancouver, Canada, May 2018. Trade-offs between performance, reliability, and energy consumption [pdf].
• Clusters, Clouds, and Data for Scientific Computing, La Maison des Contes, France, September 2016. Resilient application co-scheduling with processor redistribution [pdf].
• PASA Workshop, at ICPP'16, Philadelphia, USA, August 2016. A different re-execution speed can help [pdf].
• Int. Conf. on High Performance Computing (HiPC), Bangalore, India, December 2015. Which verification for soft error detection? [pdf].
• 6th Int. Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS15), at SC'15, Austin, USA, November 2015. Two-level checkpointing and partial verifications for linear task graphs [pdf].
• International Workshop For International Collaboration On Trustworthy Software, Rennes, France, October 2015 [pdf].
• Algorithms and Scheduling Techniques to Manage Resilience and Power Consumption in Distributed Systems, Dagstuhl Seminar, Germany, July 2015. Which verification for soft error detection? [pdf].
• 9th Scheduling for Large Scale Systems Workshop, Lyon, France, July 2014. Co-scheduling algorithms for high-throughput workload execution [pdf].
• Invited talk at the PhD Forum, held with the International Parallel and Distributed Processing Symposium (IPDPS), Phoenix, AZ, USA, May 2014. Resilient and energy-aware algorithms [pdf].
• Invited talk at the 2nd International Workshop on Power-aware Algorithms, Systems, and Architectures (PASA), in conjunction with ICPP, Lyon, France, October 2013. Energy-efficient scheduling [pdf].
• Algorithms and Scheduling Techniques for Exascale Systems, Dagstuhl Seminar, Germany, September 2013. Energy-efficient scheduling [pdf].
• 4th International Green Computing Conference (IGCC), Arlington, VA, USA, June 2013. Energy-aware checkpointing of divisible tasks with soft or hard deadlines [pdf].
• 9th Workshop of the INRIA-Illinois-ANL Joint Lab on Petascale Computing, Valpré, Ecully, France, June 2013. Energy-efficient scheduling [pdf].
• Clusters, Clouds, and Data for Scientific Computing, La Maison des Contes, France, September 2012. Energy-aware mappings of series-parallel workflows onto chip multiprocessors [pdf].
• International Research Workshop on Advanced High Performance Computing Systems, Cetraro, Italy, June 2011. Energy-aware mappings of series-parallel workflows onto chip multiprocessors [pdf].
• GT-ETIC: Groupe de travail consommation d'energie dans les TIC, Paris, France, June 2011. Performance and energy optimization of concurrent pipelined applications [pdf].
• SBAC-PAD 2010 conference, Petropolis, RJ, Brazil, October 2010. Mapping pipelined applications with replication to increase throughput and reliability [pdf].
• New challenges in scheduling theory, Villa Clythia, Fréjus, France, September 2010. Performance and energy optimization of concurrent pipelined applications [pdf].
• Clusters, Clouds, and Grids for Scientific Computing, Flat Rock, NC, USA, September 2010. Performance and energy optimization of concurrent pipelined applications [pdf].
• Euro-Par'2009 conference, Delft, The Netherlands, August 2009. Comparison of Access Policies for Replica Placement in Tree Networks [pdf].
• HeteroPar'09 workshop, in conjuntion with Euro-Par'2009, Delft, The Netherlands, August 2009. Static worksharing strategies for heterogeneous computers with unrecoverable failures [pdf]. Resource allocation for multiple concurrent in-network stream-processing applications [pdf].
• Defense of my "Habilitation à diriger des recherches", Lyon, France, July 2009. Scheduling pipelined applications: models, algorithms and complexity [pdf].
• ASTEC meeting in Les Plantiers, France, June 2009. Scheduling pipelined applications: models, algorithms and complexity [pdf].
• APDCM workshop, in conjunction with IEEE IPDPS 2009, Roma, Italy, May 2009. Filter placement on a pipelined architecture [pdf].
• Scheduling for large-scale systems, Knoxville, USA, May 2009. Mapping filtering streaming applications with communication costs [pdf].
• ALEAE kick-off meeting, Paris, France, April 2009. Algorithms to handle uncertainties [pdf].
• Gotha/MAO meeting, Paris, France, January 2009. On the complexity of multi-criteria scheduling problems for workflow applications [pdf].
• CS Colloquium, University of Memphis, TN, USA, September 2008. Multi-criteria scheduling of workflow applications [pdf].
• Clusters and Computational Grids for Scientific Computing, Flat Rock, NC, USA, September 2008. Multi-criteria scheduling of workflow applications [pdf].
• University of Passau seminar, Passau, Germany, June 2008. Scheduling pipeline workflows to optimize throughput, latency and reliability [pdf].
• 2nd "Scheduling in Aussois" Workshop, Aussois, France, May 2008. Mapping skeleton workflows onto heterogeneous platforms to optimize throughput, latency and reliability [pdf].
• APDCM workshop, in conjunction with IEEE IPDPS 2008, Miami, USA, April 2008. Fault Tolerant Scheduling of Precedence Task Graphs on Heterogeneous Platforms [pdf].
• Emory University seminar, Atlanta, USA, March 2008. Scheduling pipeline workflows to optimize throughput, latency and reliability [pdf].
• Alpage meeting, Bordeaux, France, January 2008. Scheduling pipeline workflows to optimize throughput, latency and reliability [pdf].
• LACL seminar, University of Paris 12, France, December 2007. Mapping skeleton workflows onto heterogeneous platforms [pdf].
• HeteroPar'07 workshop, in conjuntion with IEEE Cluster 2007, Austin, Texas, USA, September 2007. Complexity results for throughput and latency optimization of replicated and data-parallel workflows [pdf]. Multi-criteria scheduling of pipeline workflows [pdf].
• Workshop on Parallelism Oblivious Programming POP 2007, University of Tokyo, Japan, July 2007. Mapping skeleton workflows onto heterogeneous platforms [pdf].
• Workshop on Scheduling WS 2007, Cetraro, Italy, June 2007. Mapping skeleton workflows onto heterogeneous platforms [pdf].
• Alpage meeting, Lyon, France, June 2007. Complexity results for throughput and latency optimization of replicated and data-parallel workflows [pdf].
• ICCS conference, Beijing, China, May 2007. Mapping pipeline skeletons onto heterogeneous platforms [pdf]. Impact of QoS on Replica Placement in Tree Networks [pdf].