GPU-Accelerated Analytics and Machine Learning Ecosystems (Analytics Partners Showcase sponsored by NVIDIA)
Jim McHugh (NVIDIA)
Tim Delisle (Datalogue)
Alen Capalik (Fasdat.io)
SriSatish Ambati (H20.ai)
Location: 1E15
Join in to learn how GPU-accelerated deep learning data prep solution Datalogue can cut down 50 days of workflow to 5 days for better customer insights and greater data ROI; how GPU-native data processing solution FASTDATA.io can process a terabyte of data on a single GPU cloud in 35 seconds, instead of nine hours on Apache Spark; and how GPU-accelerated machine learning platform H2O.ai provides algorithms up to 30x faster than traditional algorithms, along with recipes for automatic feature engineering for faster and more scalable inferencing.
Datalogue
Datalogue is a NVIDIA GPU AI-augmented data operations platform. It enables automated ingestion of every data source, every data type across your entire network. Data scientists and data operators can now spend 5% or less time preparing and 95% doing data science.
FastData
Exploiting the massive parallel processing power of the GPU, FASTDATA.io has created the fastest stream processing engine in the world today. Plasma Engine™ is the first GPU-native software to fully leverage NVIDIA GPUs and Apache Arrow for real-time processing of infinite data in motion. By porting Apache Spark to Plasma Engine, we empower you to process your existing Spark workloads faster than ever on GPUs without the hassle of changing your code.
H20.ai
H2O.ai provides open source GPU-accelerated machine learning algorithms and Driverless AI platform for accurate and interpretable model building. H2O.ai brings machine learning solutions to financial services, insurance, and healthcare industries.
Learn more about Analytics partners.