Timezone isn't accessible, please provide the correct parameters
eventFeedUrl=http://realintelligence.com/customers/expos/00Do0000000aAt2/FMS_xmlcreator/a0J1J00001H0ji2_specific-event-list.xml
trackCategory=Session
eventID=a0J1J00001H0ji2
timezone=
duration=PTH
, NaNth
8:30-10:50 AM
AIML-301-1:Using AI/ML for Flash Performance Scaling, Part 1 (AI/Machine Learning Track)
Paper Title: Managing Massive Input Data in Flash for AI and Deep Learning Applications

Paper Abstract: Artificial Intelligence and deep learning solutions have ability to improve many areas of life and analyze massive amounts of information in real time. The data is often received through thousands of sensors, probes, video feeds, application data, security logs, media, GPS data… and many more sources. One of the challenges is to being able to keep up with the data ingress, especially from large number of IoT devices which can be tens of thousands in number, managing and organizing received data into data lakes which can be used for creating AI training sets, and using real time data for AI inference applications. This paper examines the types of data which can be used for AI applications, methods to optimize data ingress, organizing and preparing data into AI training sets and solutions for running AI models with inference using real-time data, and exploring possibilities that flash and storage class memory can offer in this area.

Paper Author: Dejan Kocic, Sr Systems Engineer, NetApp

Author Bio: Dejan Kocic holds MBA, MSIT degrees and numerous industry certifications. He has over 18 year of industry experience in architectural, managerial and pre-sales roles. In his current position for Netapp, Dejan is working on creating high performance computing, artificial intelligence, neural networks, deep learning, cloud and data storage solutions for NIH scientists to find cure for the world's deadliest diseases.