NERSC Data Day, Feb 21-22 2024

NERSC is hosting Data Day, a hybrid two day event on February 21-22, 2024. Join us for exciting talks and demonstrations designed to showcase the latest and greatest cutting-edge data-focused tools for scientific computing on Perlmutter. The presentations are set to occur in a fully hybrid format at NERSC, featuring speakers onsite. We strongly encourage participation both in-person and remote.

This event is geared towards HPC users of all experience levels, but will focus on intermediate and advanced topics which will help data and AI workloads run performantly at scale on Perlmutter.  First time HPC users may want to attend or review our recent New User Training and Updated Best Practices on Perlmutter before attending Data Day.  We encourage attendance from graduate students, postdocs and others who have an active role in development. The sessions will focus on tutorials covering a wide range of topics including Python, containers, workflow tools, data transfer, deep learning, and more. The emphasis is on leveraging these tools to accomplish scientific objectives across various applications in NERSC. Following the sessions, participants will have exclusive access to NERSC experts for personalized guidance on implementing these tools into their workloads. Additionally, hands-on/demo materials from the talks will be available, enhancing the practical learning experience.

After this event, NERSC will host a Perlmutter and Data Day Office Hours for the New User Training and Updated Best Practices on Perlmutter and the Data Day event on February 23, 2024 from 9 a.m. - 12 p.m. Pacific Daylight Time (PDT).

Agenda

Day 1

Time (PDT)	Topic	Presenters
9:00 - 9:10 a.m.	Introduction/Welcome
9:10 - 9:30 a.m.	Directions in Data at NERSC	Wahid Bhimji
9:30 - 10:00 a.m.	HPC Containers, Podman-HPC, Shifter	Adam Lavely
10:00 - 10:20 a.m.	Science Applications: Checkpoint/Restart in Containers	Madan Timalsina
10:20 - 10:30 a.m.	Break
10:30 - 11:00 a.m.	Choosing the right storage for your data	Steve Leak, Ravi Cheema
11:00 - 11:30 a.m.	Distributed Python at NERSC	Daniel Margala
11:30 - 12:00 a.m.	HPC-Friendly Workflows in Julia	Johannes Blaschke
12:00 - 1:30 p.m.	Lunch
1:30 - 2:00 p.m.	Data Transfers and the Globus Ecosystem	Nick Tyler
2:00 - 2:30 p.m.	Deep Learning at Scale	Peter Harrington
2:30 - 3:00 p.m.	Nvidia Triton Demo: Incorporating AI inference into workflows	Andrew Naylor

Day 2

Time (PDT)	Topic	Presenters
9:00 - 9:30 a.m.	Integrated Research Infrastructure and N10	Debbie Bard
9:30 - 10:00 a.m.	Superfacility API	Bjoern Enders
10:00 - 10:30 a.m.	Live Streaming of Large Electron Microscope Data to NERSC	Sam Welborn, Peter Ercius
10:30 - 10:40 a.m.	Break
10:40 - 11:10 a.m.	Data Stream Processing & Integrated Research Infrastructure Across Facilities	Jeng-Yuan Tsai
11:10 - 11:40 a.m.	Distributed Workflow Management using JAWS	Daniela Cassol
11:40 - 12:10 p.m.	Large Scale Transformer-based Weather Prediction	Jared Willard

Videos

The talks are available on the NERSC YouTube channel as a playlist.