Reducing cloud costs and improving performance for our nosql needs
September 26, 2018
Reducing cloud costs and improving performance for our nosql needs
Score 7 out of 10
Vetted Review
Verified User
Overall Satisfaction with Couchbase Data Platform
We are evaluating and PoC'ing couchbase to move our existing unstructured data into a more cost-effective and scalable nosql datastore. We're a small company so it's being used by our platform engineering team, and regarding business problems, as i mentioned it's mostly to reduce cost and allow for better analytics/insights. Our current nosql store doesn't really allow any kind of live querying without impacting production performance.
Pros
- Analytics (in the beta version) - being able to run broad queries and derive insights without affecting the operational store (ie. production systems)
- Performance - the memory caching is a big boost from the other nosql dbs we evaluated
- XDCR - super easy for cross region replication
Cons
- Needs a managed cloud offering (ie. no worrying about infrastructure, kubernetes or anything.....just provision and go, and no devops required)
- Documentation online could be more consolidated, and more of it :-)
- It's not immediately clear which data to combine into the same bucket, versus which data to put into their own buckets (ie. guidance for which "type"s go together)
- Better performance for our IoT processing pipeline
- Cheaper for same data/performance than existing nosql managed cloud db
- More devops cost due to no managed cloud offering (next best thing is our own kubernetes)
Cheaper, and more performant (but we have to host it ourselves).
Comments
Please log in to join the conversation