SQL Server 2014 – In Memory OLTP (Project Hekaton)

Have you been wondering what SQL Server 2014 has in store for you? Memory optimized tables are in, slow running transactions are out! With multi-version optimistic concurrency, a completely new data structure optimized for memory and a brand new set of memory optimized indexes in the form of Hash indexes and Range indexes, SQL Server 2014 comes with a bang.

With the introduction of In Memory OLTP, SQL Server 2014 also introduces a new type of compiled stored procedures. These procedures are literally compiled to machine code and run in memory accessing memory optimized tables in an efficient manner.

In-memory OLTP SQL Server 2014

In-memory OLTP Architecture

Where SQL Server really excels in its delivery of Memory optimized tables is that it really doesn't change much of how you're used to interacting with SQL Server. For the most part, the introduction of memory optimized tables can a seamless operation to your applications. You can be the hero without causing too much pain. There are a few restrictions you'll need to take care of but as you can see from the image below, T-SQL and client applications work as expected and see your tables just as any other table in the system.

Check out our new Pluralsight course on Optimizing OLTP and Data Warehousing with SQL Server 2014 to learn more about how you can utilize memory optimized tables for OLTP and Data Warehousing type applications. The course covers design principles, index types, examples of loading data, pitfalls and best practices along with how to use natively-compiled stored procedures. In the next post we'll discuss how Clustered ColumnStore indexes in SQL Server 2014 can help with your Data Warehousing workloads. Also part of the same course. Happy coding!



Leave a Comment

You must be logged in to post a comment.