Redis is the DB you didn't know you need.
April 19, 2019

Redis is the DB you didn't know you need.

Roberto Luna Rojas | TrustRadius Reviewer
Score 10 out of 10
Vetted Review
Verified User

Overall Satisfaction with Redis

We are starting to move towards using Redis as our primary storage for performance and simplicity reasons. We have been using it for the last 5 years as a Caching and Session storage mechanism, but last year we started to implement it across multiple services.
  • High Performance. Compared to other similar technologies Redis is blazing fast.
  • Built-in Data Structures. Redis facilitates the creation of Counts, Leaderboards, & Logs.
  • Scale Out. Other technologies claim to do Active-Active replication but while rebalancing, the Cluster goes irresponsive. With Redis, this does not happen and requests are still being served.
  • Management console still not quite polished.
  • No built-in GUI for Debugging keys and values.
  • Pricing model.
  • Reduced cost over other technologies but still not cheap enough to make an impact.
  • No easy way to look at the raw data from a GUI. CLI access is required to debug which is not always available for developers.
  • Dramatically reduced the response times for a positive impact on users interactions.
Yes - We replaced Couchbase because it doesn't meet the expectations. It is still a good product but it doesn't pay off the cost. It is still probably the best NoSQL for Document store which ReJSON is starting to solve but far from covering the Query-like support that Couchbase has.

ElasticCache still is a very good way to introduce the usage of Redis in an automatically managed fashion but not good enough for high performance with a reduced number of servers. Maybe this is by design so you have to buy more nodes.
  • Product Features
  • Product Usability
  • Product Reputation
  • Analyst Reports
  • Third-party Reviews
Features and performance are what makes Redis the best choice for in-memory database storage. Pricing has to be revisited at some point.
Couchbase doesn't keep up with what they offer and what really does.
MongoDB just doesn't scale out, reads are performed across multiple nodes but writes still go to the single master.
DynamoDB is good overall but just way too expensive.
Redis is well suited for applications with well-defined data usage within the NoSQL space, that is counters, queues, leaderboards, time-based, key-value hashes, or for anything that requires many secondary indexes. Still, for a tabular view, a relational DB would make more sense. In the case of full-blown JSON lookup, maybe other NoSQLs could perform as well as Redis.

Redis™* Feature Ratings

Performance
10
Availability
9
Concurrency
9
Security
5
Scalability
9
Data model flexibility
10
Deployment model flexibility
5