Working with In-Memory Tables

Abstract

In Chapter 7 you were introduced to how to build an in-memory table and the choice of either using schema-only storage or also storing the data on disk. The aim of this chapter is to move forward and look at how to work with that data outside of stored procedures, as well as at the performance improvements in-memory-based data can provide. You will also see how in-memory tables use a concurrency model that tries to ensure that processes work as quickly as possible. You’ll look at the overheads this model provides and compare it to a model that locks data so that only one process can work with a row of data. You will also look at how to merge data from a source location to a target location.