Architecture and Maintenance of VLDBs (Part 1) by Dmitri Korotkevitch

Session Title: Architecture and Maintenance of VLDBs (Part 1)
Speaker: Dmitri Korotkevitch

Abstract: The modern Data Management systems deal with enormous amount of information. The business and compliance requirements force them to collect and process large amount of data and retain them for prolonged amount of time. As result, the size of the databases grows exponentially year after year, which leads to database administration and performance challenges, and increases hardware cost. This, 2-part session, discusses various aspects of architecture, implementation and maintenance of Very Large SQL Server Databases (VLDB) working under mixed (OLTP + DW/Reporting) workload. It will:
Provide an overview of three different technologies supported by SQL Server – row-based, column-based and In-Memory OLTP and discuss their benefits, downsides and the best use-cases
Show how to design the systems in the way, that would benefit from all those technologies
Discuss typical challenges related to VLDB maintenance, such as how to reduce database backup size and hardware cost, improve RTO, maintain indexes, among with several other topics.