Director of HPC, Voltron Data. US
“Saying the quiet part out loud about data systems“
Abstract
While the HPC community has been busy pushing the flops cross over the exa barrier, we’ve done a poor job at building systems that lower data friction. One could even say that our community has accepted locked-up data as a fact of life. Many datasets are either format or vendor-locked, and we spend an uncomfortable amount of time creating glue-code to transfer and make data useful. Not only is it difficult for tools and systems to communicate with each other, but once data does arrive at its destination we still have a lot of work to do to make that data useful in HPC, and AI. Luckily we have the opportunity to unite HPC and AI if only data standards existed and data systems could communicate over a unified protocol. This talk will explore what a world without data friction looks like if we are willing to band together and agree over data standards.
Bio
Fernanda holds a Master’s in Materials Science Engineering from the University of Florida and a Bachelor’s in Physics from Florida International University. She’s passionate about supercomputers and finding ways to improve efficiency, utilization and user friendly interfaces in the scientific and data science communities.
She is currently a Director at a startup, Voltron Data Inc, and leads both developer relations and leads R&D partnership efforts to develop better tools for data analytics.
Her professional background is broad but always with a focus on creating a better user experience. Her experience includes:
NextSilicon (silicon startup): Director of Applications whose primary responsibility was leading a team testing and defining performance metrics for their first accelerator.
BioTeam (consultancy firm): Senior Scientific Consultant where during the pandemic she engineered systems for pharma companies to analyze genomic and other data.
NVIDIA (AI hardware platform): GPU Developer Advocate for Healthcare HPC+AI primarily responsible for life sciences. She led the acquisition of Parabricks, a GPU accelerated alternative to GATK.
Oak Ridge National Laboratory: She spent 6 years as HPC User Assistance Group as an HPC Programmer and Training Coordinator at the Oak Ridge Leadership Computing Facility. She participated in the CORAL project that selected Summit as the next supercomputer to replace Titan. She was also part of the Exascale Computing Project and was co_PI on two ECP projects: Kokkos and Training. For her last two years, she was an HPC Data Scientist working on the CANDLE project (aka Biden’s Cancer Moonshot). Throughout her time at ORNL she also served on the C++, OpenMP, and OpenACC language committees where she was the release manager for the latter.
Genus Plc: She was an HPC Systems Manager and Scientific Programmer. Her major role was to support ~20 scientists worldwide. Three major projects included: building an HPC cluster, parallelizing all production codes, and lastly a columnar genotype database, to host a real time searchable database, containing all of the panels and deep sequencing of thousands of samples. The latter led to a neat imputation project that saved Genus a lot of money in sequencing costs and could accurately predict sequences for low-density genotyped animals from strategic high-density genotyped animals.
Originally born in Rio de Janeiro, Brazil, she moved to Florida at a young age and enjoys sunshine, tennis, home projects, and her family. She plans to learn to sail someday.
Please click to download 2024 Slides as pdf.
multicore.world 