Interesting news from DO:
Have not really looked into the details about what their value proposition might be against Mongo’s own Atlas. But competition is certainly beneficial for us all.
1 Like
This is quite interesting, Mongo changed their license agreement a few years back to prevent this (at least on their community edition) - pretty much because of AWS’ DocumentDB. Nice to see some alternatives are still being allowed through - pretty competitive pricing too from what I can see.
According to the blog post, DO’s service is “built in partnership with and certified by MongoDB Inc.” So I guess it all comes down to whether or not you’re willing to share revenue with MongoDB Inc or not. DO is and AWS was not.
Although, if I recall correctly, one issue that Mongo cited when explaining their license change was that AWS was using their product to build out DocumentDB, but did not contribute their own new code back upstream to Mongo. So perhaps this was a reason as well.
2 Likes
That’s why I’m surprised their pricing is so competitive, if they’re kicking some back up to Mongo, where Atlas (presumably) does not have to.
Question is, does it provide Oplog?
1 GB with 2 replica costs 45 USD while 2GB on Atlas costs 60 USD. It is true, in DigitalOcean you get an option for 1GB but I do not see how the pricing is competitive.
1 Like
I guess it depends on how much the 2 gb on do is
The main difference between MongoDB and DO is that with MongoDB you also pay for data transfer and storage!. Example:
M10
10 GB
2 GB
2 vCPUs
$0.08/hr
looks like it will cost you $57.6/month, but that amount will at the very least double.
So that will probably make DO about 30% cheaper
2 Likes
One thing I’ve been interested in is full text search abilities with Atlas Search. But, I’m pretty sure it’s not included or free. Anybody have any experience with it?
It also seems there are some limits of Digital Ocean’s MongoDB implementation. Like: D.O. currently only supports MongoDB v4.4.
We use Atlas Search and it’s pretty cool. Also, it’s free and included on all tiers AFAIK. I haven’t tweaked our use of it since we started so not sure how much it’s advanced since then but it’s much better and more configurable than the basic text search and is great for use cases where your customers might be searching over multiple fields and you want to create a scoring system.
Also - I’m just noticing the previous message in the thread about the price doubling with transfer/storage costs. I guess that could happen if your DB is really big but it’s not our experience.
1 Like
We wanted to try it out as well, but the fact that’s it’s impossible to use it locally was a no-go for us. Instead, we did setup an Elasticsearch cluster (OK, OpenSearch) and used monstache to synchronize MongoDB collections to Elasticsearch indexes. The entire setup is extremely cheap and effective — it costs less than 1% of the entire setup (servers + database).
2 Likes
Another cash grab. Just run mongo locally until you reach 40k+ daily active users. I run it on a $10 droplet. Doesn’t even sweat.
1 Like