11-08, 10:10–10:50 (US/Eastern), Music Box
NetworkX is easily the most popular graph analytics library available, not just among other graph analytics libraries in Python, but quite possibly any language. With monthly downloads in the tens of millions, a large and friendly community, a wealth of documentation and examples, and a wide breadth of algorithms behind an easy-to-use API, NetworkX has earned its household name status among data scientists. Despite these advantages, NetworkX has struggled to keep up with the demands of many modern workloads resulting from our world's growing appetite for large data, which forces users to rewrite their code for a harder-to-use but more performant library.
However, NetworkX recently added support for third party backends to accelerate its popular API, and NVIDIA has responded with nx-cugraph, a NetworkX backend that uses cuGraph for extremely fast and scalable graph analytics. This talk will provide the background and details covering the features that allow NetworkX to leverage third-party backends such as nx-cugraph, and demonstrate how users can run their NetworkX code on CPU-only systems, then without code modifications, migrate to a GPU-enabled environment to realize huge performance increases.
NetworkX is arguably the most popular graph analytics library available today; it's extremely easy to install and use, supports a huge assortment of popular analytics, and has excellent support via its community and documentation. But despite these benefits, many users are forced to move on to another graph analytics library when the limitations of NetworkX's pure-python implementation prevents them from using it for larger workloads.
Will NetworkX users ever be able to have the ubiquity and ease-of-use they've grown accustomed to without the limitations of a pure-python implementation?
The answer is yes! Graph analytics that are fast and scalable but still easy to use and well-documented are possible simply by using NetworkX with an accelerated backend. The backends available to NetworkX today enable support for running analytics on GPUs, leveraging linear algebra through graphBLAS, incorporating graph databases, and more, without requiring users to change their existing NetworkX code. This changes the state of graph analytics by making it accessible to more users and more use cases than ever before.
Attend this talk to hear just how easy it is to use NetworkX to unlock the hidden secrets in your large, real-world data that only fast graph analytics can reveal. Detailed background knowledge of NetworkX and graph analytics is not required.
An approximate time breakdown is as follows:
8 mins - Brief introduction to graph analytics and NetworkX, why it's the most popular graph analytics library, and what its limitations are.
15 mins - How do accelerated backends work, what's available, and how can they make NetworkX work for workloads both big and small?
7 mins - Q&A
No previous knowledge expected
Rick Ratzel is an engineering manager for RAPIDS cuGraph - a library of GPU-accelerated graph algorithms. Rick joined NVIDIA in January 2019, bringing several years of experience as a technical lead for teams in industries that include test and measurement, electronic design automation, and scientific computing.