See a map of all the places I've given a talk (based on the list below)
Check the list of talks per topic
Complete list of talks since 2023:
Invited Talk, SIAM CSE25
Fort Worth, Texas
I had the pleasure of giving an invited talk at the SIAM CSE25 workshop: Neural Acceleration, Surrogate Models, and Learning Techniques for HPC Kernels. The minisymposium offered an introduction into current solutions for approximating expensive kernels with heuristics based on machine learning and the trade-offs encountered when used in HPC environments.
Talk, ADIOS User Group Meeting
Oxford, UK
The ADIOS User Group Meeting was a one day event attended by users from the US and Europe and has included talks about success stories using ADIOS in extreme scale storage I/O, in situ data processing and/or code coupling, as well as their newest requirements and requests to the ADIOS team. In the second half of the day ADIOS developers presented latest features like campaign management, remote data access, and the design of schemas independent from file formats to gather ideas and feedback from the users.
Talk, Open Data Schema Workshop
Oxford, UK
The Open Data Schema Workshop held in conjunction with the Data-driven Plasma Science and Engineering Workshop in Oxford, UK takes a first step at establishing consensus dor data formats across studies, experiments and simulations. The workshop gathers domain scientists and computer scientists from Europe and the US and provides an understanding of the I/O requirements of advanced HPC simulations and of the challenges unique to specific configurations(e.g. disruption avoidance and mitigation in tokamaks).
Invited Talk, 18th Scheduling for large-scale systems workshop
Montréal, Québec, Canada
I will be giving a talk at the 18th scheduling workshop workshop, which will take place at the École de Technologie Supérieure in Montréal, Québec, Canada, from Tuesday, July 8, to Thursday, July 10 (2025).
Talk, Kokkos tea-time
Knoxville, Tennessee
Scientific data analysis often involves complex queries across distributed datasets, requiring manipulation of multiple variables and generating derived data on the fly. Derived quantities are obtained by mathematical transformations of primary data (generated directly by applications) and allow researchers to focus in their analysis on specific aspects of their simulation. For example, in combustion simulations, calculating the magnitude of the velocity (primary data) creates a derived variable that effectively identifies areas of high interest, such as regions with intense burning. In this talk I will present optimization opportunities when the decision on when to compute the derived variables is offloaded to the I/O library and how Kokkos can be used to handled them efficiently for I/O intensive applications.
Paper Talk, IEEE 36th SBAC-PAD
Hilo, Hawaii
Online presentation for the IEEE 36th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD) conference. ORNL scientists explored methods for managing the amount of data produced by HPC simulations and retrieving needed data more efficiently. The researchers found that offloading this task to an input/output (I/O) library would allow them to choose the best strategy in each scenario since all the information necessary to choose between the trade-off is present at this level.
Tutorial, IEEE eScience
Osaka, Japan
This half-day tutorial provides a comprehensive introduction to building blocks of complex scientific workflows, including I/O, remote data access, and visualization. The tutorial features live demonstrations involving a realistic simulation, data analysis and visualization, which illustrate how to use these tools in a plug-and-play manner. Attendees will see a workflow composed of simulation, data analysis and data visualization and run in various configurations of file-based and in situ setups.
Tutorial, Maison de la Simulation, CEA
Paris, France
Half-day tutorial as part of the CEA event Parallel IO and in situ analytics: High-performance data handling @ Exascale that showcases the must-know tools for high-performance input-output on supercomputers and discover the latest tools for in situ data analytics. The event included presentation for ADIOS, HDF, NetCDF and PDI and interactions with developers of the leading tools in their category as well as IO administrators of the top French national super-computing facility.
Invited Talk, 17th Scheduling for large-scale systems workshop
Aussois, France
My presentation focuses on efficient strategies for querying large datasets, specifically addressing quantities of interest and derived data. Although not directly about scheduling, these strategies create analysis tasks and data transformation needs that can strain study resources, especially at scale. To fully realize the potential of these data management and query techniques, and to ensure timely completion of analyses, the development and implementation of custom scheduling solutions will be essential.
Invited Talk, NeurIPS'23: Workshop on Advancing Neural Network Training
New Orleans, Louisiana
The first edition of the Workshop on Advancing Neural Network Training (WANT) held as part of thre NeurIPS workshop presents talk relevant to computationally efficient training, which covers both memory-related issues and scalability issues, scalability and resource optimization. My invited talk explores the performance trade-offs within the multilayer high-performance I/O systems under the strain of emerging AI applications (inluding deep neural networks and large language models).
Talk, Kokkos User Group Meeting
Albuquerque, New Mexico
My talk at the 2023 Kokkos User Group meeting presented early performance results on integrating Kokkos into the ADIOS I/O library.
Tutorial, The International Conference for High Performance Computing, Networking, Storage, and Analysis
Denver, Colorado
Full-day tutorial at SC23 focuses on building in situ and file-based data processing workflows for extreme scale systems, including interactive, on-demand, in situ visualization of the data, and including performance profiling of the entire workflow.
Invited Talk, Science and Technology Facilities Council (STFC)
Harwell, UK
I had the pleasure of being invited to give a talk to the Scientific Machine Learning Research Group at the Rutherford Appleton Laboratory. My talk focuses on performance considerations and trade-offs of emerging AI workflows from the I/O sub-system perspective.
Invited Talk, Schloss Dagstuhl - Leibniz Center for Informatics
Wadern, Germany
The Dagstuhl Seminar on Integrating HPC, AI, and Workflows for Scientific Data Analysis held between Aug 27 and Sep 1st 2023 focused on understanding of AI+HPC integrated workflows, elaborating on the different modes in which AI and HPC components could be coupled within workflows and the corresponding paradigm shift of HPC systems towards real-time interaction within workflows. My talk focused on profiling the I/O patterns in applications running on HPC throughout the years and identifying limitations on scaling modern applications that combine HPC and AI.
Invited Talk, 16th Scheduling for large-scale systems workshop
Knoxville, Tennessee
My talk focuses on ways to build and automate AI workflows by separating the data and computational planes and attaching tasks to data offloading the I/O management to specialized processes. Experiments done on a whole slide image processing workflow for cancer research shows that the framework can leverage the data access and pre-processing patterns in order to fetch in a more efficient way the required input data in the needed format and ordered by the needs of each application within the workflow.