Overall Satisfaction with Riak
Riak is used as the main database/K-V Store for the company's product, the Dittach Platform. It is used to store information on all the objects/documents we manage in our environment. We selected Riak because it is highly available, highly scalable, and has Apache SOLR built into the data store for fast searching and indexing.
- Highly available: If nodes go offline for any reason, the system still operates.
- Highly scalable: There is a minimum of 5 nodes, which can handle a lot by themselves. When scaling is required, it can be done easily, with minimal to no downtime on large scales.
- Very fast searching: Riak has SOLR indexing built-into the core product, which makes querying for data very fast.
- When the index definition changes, reindexing takes an extremely long time.
- Support (both paid and community based) is very very lacking.
- It is expensive to run.
- Riak has had a positive impact in the fact that it is still a very fast data store and can scale very well over time.
- Riak has had a negative impact as it is one of the most expensive pieces of our architecture.
- Riak's SOLR engine failed us at a critical moment which seriously hurt the company. Our use case is very different than the norm, so I don't expect this to affect most people.
Every database has positives and negatives. Redis is very good at set operations, but not as good at executing fast queries. MongoDB is a much cheaper data store than Riak, but not as highly available and scalable. Searching Riak is also faster than searching MongoDB or Redis. Riak also scales better than MongoDB in the long run.