In-Memory Queries Explained
Traditionally relational databases use disk to process data for a query operation. In effect, this meant that if a query was run against a large database, the data that needed to be pulled from the database in order to produce the query result was stored on disk and processed from disk. When having to process millions of rows of data, the time taken to access data from disk has a significant impact and producing results can take hours or even days.
In-memory queries are a way of taking all of the data required to perform a query and placing it into memory (RAM). Doing this can significantly speed up query processing time, leading to faster results, quicker decisions and less time with database managers sitting and waiting for queries to process.
Many of the existing database companies now offer in-memory databases. Companies such as GridGain offer a platform to enable even traditional disk-based databases to process queries in memory – this would be an example of an in-memory query.
In-memory queries deal with two specific issues. Technically, they enable faster processing as data volumes become massive. Commercially, they allow decisions and answers to be derived in a timescale that meets the demands of today’s instant response business and customer expectations.