Back to the Future: Teradata -> Redshift -> Teradata
Nordstrom took a three-year journey from Teradata to Redshift and back to Teradata. We will discuss why we migrated to Redshift, the positives and negatives and then why we are migrating back to Teradata for Enterprise Analytics.
Breakout #1752
The Future of Analytics
This talk examines key trends in analytics deployment and developments in advanced technology, along with the implications of each. Specific areas of focus include: (1) data acquisition and delivery, (2) operational intelligence in the real-time enterprise, and (3) analytic applications architecture.
Breakout #1808
Turning A Resource Problem Into A Goldmine
Rakuten analysts and data scientists faced a crisis when exponentially growing demand for aggregated sales data outgrew the processing capacity of the existing Teradata installation, and resource competition triggered chronic timeouts across the business. This presentation tracks the company’s journey, the new resources built, and how the resulting commitment to best practices led to productivity gains and a lucrative data monetization opportunity.
Breakout #1858
Hybrid Cloud, Multi-Cloud: All the As-a-Service Options for Vantage
Most enterprises surveyed expect to have a hybrid analytics environment that consists of both on-premises and cloud. Why? Because data gravity exists and forklifting your entire data center into the cloud carries both cost and risk. Learn how Teradata customers use hybrid and multi-cloud architectures to optimize performance and price for maximum value from each deployment type while keeping future evolution options open.
Breakout #1736
US Air Force Soars in the Azure Cloud
In 2018, the US Air Force launched a data analytics program cloud migration project. The program aggressively began the migration to Azure GovCloud leveraging an agile project approach, quickly learning the ins and outs of migrating and optimizing a complex enterprise ETL, BI and Data Warehouse operation to the public cloud. This session examines the original implementation plan as well as what was actually fielded.
Breakout #1759
Not reinventing the wheel. Using Teradata Vantage ML functions with Python for quick and easy success
This case study presentation demonstrates how we leveraged Teradata Vantage and Python to make machine learning models to predict how many people will attend Utah Jazz games. We’ll cover how to structure data to work for a machine learning problem and then how to automate feature and model testing, speeding up the process of turning data into insight.
Breakout #1760
Vantage and R, a User’s Perspective: Building a Modern Analytics Workflow at Wells Fargo
Join a longtime R and Teradata Aster user to learn about the benefits and opportunities he discovered in converting a complex analytical application from Teradata Aster to Vantage. The application covers complex customer behavior data including multiple click streams, text, and administrative KVPs in order to predict customer intent.
Breakout #1806
Loblaw’s Vantage Journey
This two-part presentation documents the migration journey of Loblaw, Canada’s largest grocer, to Teradata Vantage. Part 1 covers migration preparation steps, challenges faced, and query performance gains. Part 2 covers Machine Learning Engine experiences, New SQL Engine, AppCenter, TASM workload modifications, and extending the reach of Vantage to other platforms in local and Disaster Recovery data centers using QueryGrid.
Breakout #1839
Analytics at Scale using Python and R on Vantage
This session covers three approaches to scaling, along with the advantages and disadvantages of each: using standard Python and R to extract from databases, accessing Python or R via an SQL script table operators, and accessing Vantage via Python using TeradataML or via R using tdplyr.
Breakout #1900