Presented By O’Reilly and Cloudera
Make Data Work
September 11, 2018: Training & Tutorials
September 12–13, 2018: Keynotes & Sessions
New York, NY

Accelerating financial data science workflows with GPUs

Joshua Patterson (NVIDIA), Onur Yilmaz (NVIDIA)
5:25pm–6:05pm Wednesday, 09/12/2018
Data science and machine learning
Location: 1A 15/16 Level: Intermediate
Secondary topics:  Financial Services

Who is this presentation for?

  • Data engineers, data scientists, quantitative analysts, and financial analysts

Prerequisite knowledge

  • A basic understanding of pandas, Numba, and scikit-learn
  • Familiarity with data science

What you'll learn

  • Understand why GPUs should be used in big data architectures
  • Explore benchmarks for GPU performance
  • Learn about GoAi and how to get involved


In order to make the most informed and profitable decisions, financial services organizations rely on using simulation methods, advanced modeling, and data mining to better model and understand the possible outcomes of their potential decisions. Due to innovations in computing power and software optimizations, organizations are able to run these highly complex operations faster than ever before, which in turn also generates and requires more data than ever before. Because this data is demanded at a staggeringly fast rate, organizations are increasingly unable to operationalize data fast enough to keep up with the requirements. Existing big data frameworks and machine learning solutions have been a great first step toward tackling this problem, but the cost to scale to the volume and the velocity of current needs has proven to be prohibitively expensive. This has led many to start adopting GPUs for data science.

Joshua Patterson and Onur Yilmaz discuss several GPU-accelerated data science tools and libraries that can be used to process financial data more efficiently. They also explore the evolution of the GPU data science ecosystem and discuss the motivation behind and value of the GPU Open Analytics Initiative (GoAi)—a collection of libraries, frameworks, and APIs built on Apache Arrow designed to simplify and accelerate development and performance. They conclude by sharing machine learning and deep learning approaches to various financial services use cases.

Photo of Joshua Patterson

Joshua Patterson


Joshua Patterson is a director of AI infrastructure at NVIDIA leading engineering for RAPIDS.AI. Previously, Josh was a White House Presidential Innovation Fellow and worked with leading experts across public sector, private sector, and academia to build a next-generation cyberdefense platform. His current passions are graph analytics, machine learning, and large-scale system design. Josh loves storytelling with data and creating interactive data visualizations. He holds a BA in economics from the University of North Carolina at Chapel Hill and an MA in economics from the University of South Carolina Moore School of Business.

Photo of Onur Yilmaz

Onur Yilmaz


Onur Yilmaz is a deep learning solution architect at NVIDIA, where he works on deep learning use cases for finance and helps researchers and data scientists adopt deep learning and GPU technology. Onur holds a PhD in computer engineering from the New Jersey Institute of Technology; his dissertation focused on traditional machine learning and high-performance signal processing for finance.