Land O’Lakes: How Free-Form Data Lakes Are Complementing Structured Data Warehouses
TDWI Speaker: Philip Russom, TDWI Research Director
Date: Wednesday, September 28, 2016
Time: 9:00 a.m. PT, 12:00 p.m. ET
Webinar Abstract
As the data warehouse environment (DWE) continues to evolve, one of its strongest trends is the diversification of data platforms. A rigorously structured relational data warehouse is still at the heart of the DWE, but it is being joined more and more by other platform types, including data platforms based on columns, appliances, graph, streaming data, and open source.
Why so many platform types? Data itself is diversifying into a wider range of structures, sources, and latencies. Meanwhile, organizations are diversifying how they leverage data via analytics, reporting, complete views, and real-time operations. With such diversity, data management teams need multiple platforms and tools so they can give each data type and use case the appropriate rich functionality.
In the thick of these changes, Hadoop has emerged as an important enabler of DWE diversification because of its ability to handle a wide range of data types, ingestion methods, and processing for analytics and integration—all with linear scalability, a relatively low cost, and solid interoperability with both old and new systems. As users gain more experience in this area, they conclude that the leading approach to organizing a data store on Hadoop is the data lake.
In this webinar, you will learn:
- What data lakes are today and where they are headed
- The many business and technology use cases a single data lake can support
- How a free-form data lake is a useful complement to a structured, relational, and dimensional data warehouse
- How a data lake starts as a free-form data store but eventually adds “just enough structure,” as required by recurring tasks in reporting and analytics
- How logical data warehouse methods and in-memory processing can get more value out of a data lake
- Miscellaneous design, architecture, governance, and maintenance issues with data lakes
- Enabling technologies and practices that make the logical data warehouse fully practical today, including semantic layers, data view technologies, data fabric, real-time functions, columnar databases, and in-memory databases and analytics processing
Philip Russom, Ph.D.