How To Run a Redis SQL Query Without Disruption

Related E-Book download: The Importance of In-Memory NoSQL Databases

Running a Redis SQL query doesn’t have to be difficult. I actually raised this point a few years ago when speaking to a friend who manages data warehousing solutions at a retail company. We began talking about Redis’ queries once he explained a problem that he was facing.

“We have a pain point with our data warehousing solutions. We have use cases where we need to record data and perform analytical operations in real-time. However, sometimes it takes minutes to get the results. Can Redis help here? Keep in mind that we can’t rip and replace our SQL-based solution all at once. We can only take a baby step at a time.”

Now, if you’re in the same situation as my friend, we have good news for you. There are a number of ways you can run a Redis query and introduce Redis into your architecture without disrupting your current SQL-based solution.

Let’s explore how you can do this. But before we go any further, we have a Redis Hackathon contestant who created his own app that allows you to query data in Redis with SQL.

Watch the video below.

Remodel your tables as Redis data structures

It’s quite straightforward to map your table to Redis data structures. The most useful data structures to follow are:

• Hash
• Sorted Set
• Set

One way that you can do this is by storing every row as a Hash with a key that’s based on the table’s primary key and have the key stored in a Set or a Sorted Set.

Figure 1 shows an example of how you could map a table into Redis data structures. In this example, we have a table called Products. Every row is mapped to a Hash data structure.

The row with primary id, 10001 will go in as a Hash with the key: product:10001. We have two Sorted Sets in this example: the first to iterate through the data set by the primary key and the second to query based on price.