UCLA-IDRE and San Diego Supercomputer Center (SDSC) are happy to announce the workshop “Advanced Computing and Data Analytics using Comet” as per the following details:

Venue: 5628 Math Sciences Bldg.
RSVP: Link to the registration form (*lunch is provided)
Requirements: Bring your laptop for hands-on exercises.
Questions: Contact T.V.Singh

Link to the presentation slides: https://github.com/sdsc-scicomp/2019-11-08-comet-workshop-ucla

Agenda:

• 9:00 AM – 9:10 AM: Welcome
• 9:10 AM – 10:10 AM: Introduction to SDSC Resources – Current and Future
  • Comet, hardware overview
  • Software, stack, queues, etc
  • Overview of upcoming Expanse system
• 10:10 AM – 10:20 AM: Short break
• 10:20 AM -12:00 PM: Python for High Performance Computing (HPC)
  • Use of Jupyter notebooks on Comet
  • Scaling using IPython parallel
  • Using Numba to run pure python codes on GPUs
  • Distributed parallel computing using Dask
• 12:00 PM – 1:00 PM: Lunch (provided)
• 1:00 PM – 2:30 PM: Introduction to Machine Learning and Deep Learning on Comet
  • Overview of machine learning / deep learning tools available on Comet.
  • Examples using R, Python, and Keras with TensorFlow.
• 2:30 PM – 2:40 PM: Short break
• 2:40 PM – 4:00 PM: Containers for Scientific and High Performance Computing (HPC)
  • Introduction to Containers – advantages and limitations
  • Overview of containers on HPC systems
  • Singularity Essentials
  • Hands-on examples showing usage of containers – CPU, GPU, and MPI cases
  • Hands-on example showing the import of Docker container
  • Info on building images
• 4:00 PM: Adjourn