I think MongoDB is great, so will you!
April 05, 2016

I think MongoDB is great, so will you!

Joshua Austill | TrustRadius Reviewer
Score 9 out of 10
Vetted Review
Verified User

Overall Satisfaction with MongoDB

We use MongoDB to store analytics information from social networking sites, and also as our main datastore for our Intranet. It works well for both by providing very fast access to our data, and in a very simple way. All of our data is consumed by ASP.NET and used mainly in JavaScript, so working with documents is a very natural fit.
  • Replication, simplest replication I've ever had to set up, and it works very well.
  • Performance, because you are simply retrieving documents it is very fast. I've seen people try to use it like a relational system and have issues, but if you learn how it is intended to be used you will have very little concern with performance in my experience.
  • Maps to objects because it's BSON. Serializing is a major strength of MongoDB to me. It is pretty awesome to just grab a document and have an object in memory and away you go!
  • .Net driver implementation, I would like to see a driver that more closely aligns with the MongoDB way. Having to use tons and tons of helper classes to build queries is kind of a pain to me.
  • Recovery, it would be great to see ways to refresh replication and sharding settings once they are broken. The current path is to start over with new nodes and restore data. That could be improved in my opinion.
  • They don't include an init script for Mongo's service, which is really a shame to me.
  • Much faster development time.
  • Price is fantastic compared to MSSQL when you consider OS costs and the entire package.
  • Only negative to me is the lack of DBA skills for it, due to it being a new player in the field. I feel like that will get better as time moves on however.
I selected MongoDB because it works for well with web interfaces. All of the RDBMS alternatives would have required a lot more time writing schemas and working around retrieving data and mapping it. That could have been somewhat mitigated with Entity Framework, but that again adds another piece of complication that I would rather just skip myself.
I am now a certified MongoDB Developer, so I see myself using it for quite some time until something better comes along for what I am doing. So far nothing has really caught my attention, but I will continue to keep looking at alternatives, things change constantly in this world.
I would say get familiar with the document model that MongoDB uses and apply it to your workload and see if it's a good fit. For most things web based I see it being a fantastic fit, the new features in 3.2 I think will make it pretty decent for analytics as well. I don't see it working great if you need your data extremely normalized and can't deal with documents however.