This event is a double header.
TECH TALK #1: "Data and ML at scale using Kubernetes"
Jayant Thomas, Director/Head of AI and Machine Learning, Veritas Technologies
In this talk, we will go over ML & Data Platform architecture used at scale using Kubernetes at the heart of compute layer and a NoSQL for managing data. We will illustrate the platform further using use-cases from Storage and Backup to compute System Reliability Score (SRS), Storage Forecasting, Configuration Drift and other algorithms used for many 100's of thousands of Storage appliances. We will analyze the pros and cons of using this architecture and close it with lessons learned.
Jayant Thomas (JT) has a passion for AI, IoT, Machine Learning and Cloud Native architectures at scale. His passion has led him to many successful adventures at Veritas, GE, Oracle, AT&T, Nuance and other startups in building platforms at scale. JT is a MBA from UC Davis along with M.Tech from NIIT, and has more than 15 patents in the IoT, NLP processing and Cloud architectures. JT is also an enthusiastic speaker/writer and contributing to many stimulating thoughts across many conferences and meetups. JT in addition is an active fitness and health freak dabbling in various diets and health fads.
Author of Best Selling "IIoT Application Development book":
TECH TALK #2: "Automated time series forecasting, backtesting, and optimization"
Marcello Tomasini, PhD, Sr. Data Scientist, Veritas Technologies
Traditionally, time series forecasting has been a popular discipline among the finance realm but often neglected as an area of machine learning. This changed dramatically with the growing popularity of IoT, sensor networks, and streaming data which lead to the collection of vast amount of time series data, often stored in ad-hoc time series databases. Forecasting is now used for a moltitude of use cases from application performance optimization to workload anomaly detection. The challenge then is to automate a process that was handcrafted for the analysis of a single data series constitued of just tens of data points to large scale processing of thousands of time series and millions of data points. In this talk, we will tackle on some of the issues and solution to deal with time series forecasting at scale, including continuos accuracy evaluation and algorithm hyperparameters optimization. As a real world example we will be discussing the solution implemented in Veritas Predictive Insight which is capable of training, evaluating and forecasting over 70,000 time series daily.
Marcello Tomasini is a "Computer Engineer and Scientist interested in Machine Learning, Computer Security, Complex Networks, and Biology with a Think Different life style”.
Marcello holds a B.S. and a M.S. in Computer Engineering from University of Modena and Reggio Emilia, Italy, and a Ph.D in Computer Science from Florida Institute of Technology, USA. He has several papers published in international peer-reviewed conferences and journals in the areas of mobile sensor networks, human mobility modeling, and machine learning. He currently works as Sr. Data Scientist at Veritas Technologies where he designed and developed the system reliability score and the storage forecasting algorithms implemented in Veritas Predictive Insight. His free time is a mix of gym/bootcamps, machine learning meetups, and traveling.