[Pycon] [new paper] "Alessandro Amici" - Meet dask and distributed: the unsung heroes of Python scientific data ecosystem.

[info a pycon.it]

Title: Meet dask and distributed: the unsung heroes of Python scientific data ecosystem.
Duration: 45 (includes Q&A)
Q&A Session: 15
Language: en
Type: Talk

Abstract: Thanks to its world-class data tools and libraries, like Numpy, Pandas, Jupyter, Matplotlib and xarray, Python is becoming the language of choice in many scientific communities from Physics to Climate Science, from Earth Observation to Economy.

A turn-key but less-know component of the scientific ecosystem is the dask library that enable seamless parallel, distributed and GPU computing in most cases without code changes.

We will show how easily simple tasks in climate science become big data problems and how mastering xarray, dask and dask.distributed turns them back into simple tasks, possibly on a large cluster of VMs (that you can easily provision from your preferred cloud provider).

Tags: [u'xarray', u'dask', u'dask.distributed', u'Jupyter', u'climate-change', u'earth-obeservation', u'Big-Data', u'pandas']