Data Science using parallel computing with Dask (@ Tech Park at Cal Poly)

Learn and experience hands on how parallelism can up your level of data science. Personal help and visualizations on how to implement Dask in your notebooks.

Cal Poly Computer Science students Ian McCurry, Jeremy Kozlowski, Evan Pomeroy, Rohit Bathula under the direction of Professor Ayaan Kazerouni will introduce Dask and the advantages of parallel computing. They will demonstrate the advantages of parallel computing as well as things to be cautious of when using it. They will walk through a live demo as well as a coding exercise for everyone to participate in. Both the instructors and teaching aids will be present to assist with the exercise.

Suggested minimum level of python experience: ability to open a jupyter notebook and run code is useful for following the hands-on exercise, but all participants are welcome.

Outline:
A. Overview of Parallel Computing (Intro, pros and cons)

B. Intro to Dask
- Show metrics of parallel computing advantages/disadvantages against different-sized datasets

• Give a prompt and show the initial run without parallel
• Implement parallelization for the class
• Run again with parallelism

C. Short live demo of using Dask

D. Give exercise: dataset and sample Python code (pr a e-dask)

Materials, including library requirements and file downloads, will be provided closer to the date on this page. Stay tuned!!!

The room does have a capacity of 30-35 attendees, so please keep your rsvp up to date.

***

The event will be held at Cal Poly, Tech Park (35.3054128,-120.673986) with ample and free parking!

If you enter Cal Poly from Hwy. 1
You are on Highland Drive
Turn LEFT on Mt. Bishop (Before the train trestle)
Turn LEFT into the Technology Park
Proceed to the lower parking lot

Enter through the double doors on the lower level. The conference room is on the left