CFD Events Calendar, Event Record #27999
How to Analyze Your Time-Dependent Data 6 Times Faster
|
TECPLOT WEBINAR - Parts of the post-processing pipeline are
inherently single-threaded - data
loading and rendering come to mind. The challenge is how to
reduce these bottlenecks to speed up
the process.
|
Date: |
April 3, 2019
|
Location: |
United States
|
Web Page: |
http://bit.ly/April2019TecplotWebinarCFDonline
|
Contact Email: |
k.anderson-deavila@tecplot.com
|
Organizer: |
Tecplot, Inc
|
Application Areas: |
Aerospace
|
Special Fields: |
High Performance Computing
|
Softwares: |
Tecplot
|
Type of Event: |
Seminar, International
|
|
Description: |
Let’s face it, analyzing large amounts of transient data
can take forever. Animating images to a movie can take an
entire weekend to complete. Why can’t Tecplot just do it
for me, and do it faster?
This Webinar will show you how the power of Python can
take advantage of multi-process parallelism. Using
Tecplot 360’s Python API, PyTecplot, you can get your
results up to six times faster.
Webinar Agenda:
1. The Problem - Post-processing time-dependent data can
take a long time and consume large amounts of RAM.
2. The Solution - Reduce memory usage and take advantage
of multi-processing – both will drastically speed up
transient post-processing.
3. The Results - Get your results up to six times faster!
4. Caveats - Of course there are caveats! We will explain
each of these: hardware constraints, licensing, dataset
compatibility.
5. Future Work - We plan to test this on a distributed
HPC system.
6. Q&A
|
|
Event record first posted on March 21, 2019, last modified on March 23, 2019
|
[Back to Events Calendar Front Page]