Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We cut our Mongo DB costs by 90% by moving to Hetzner

prosopo.io

Post reply on HN
Print view View original

101–110 of 214 posts

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#101

Post by arbol »

cpursley wrote: post #42

Why in the world do people choose Mongo over Postgres? I'm legit curious. Is it inexperience? Javascript developers who don't know backend or proper data modeling (or about jsonb)? Is this type of decision coming down from non-technical management? Are VCs telling their portfolio companies what to use so they have something to burn their funding on? It's just really confounding, especially when there's even mongo-api…

Personally I've found it faster to build using mongo cause you don't need to worry about schemas. You get 32mb per document and you can work out your downstream processing later, e.g. cleanup and serve to postgres, file, wherever. This data is a big data dump that's feeding ML models so relational stuff is not that important.

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#102

Post by notTooFarGone »

mads_quist wrote: post #57

OK guys, running on a single instance is REALLY a BAD IDEA for non-pet-projects. Really bad! Change it as fast as you can. I love Hetzner for what they offer but you will run into huge outages pretty soon. At least you need two different network zones on Hetzner and three servers. It's not hard to setup, but you need to do it.

There are so many applications the world is running on that only have one instance that is maybe backupped. Not everything has to be solved by 3 reliability engineers.

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#103

Post by canucktrash669 »

0xbadcafebee wrote: post #17

> Here's how we managed to cut our costs by 90% You could cut your MongoDB costs by 100% by not using it ;) > without sacrificing performance or reliability. You're using a single server in a single datacenter. MongoDB Atlas is deployed to VMs on 2-3 AZs. You don't have close to the same reliability. (I'm also curious why their M40 instance costs $1000, when the Pricing Calculator ( https://www.mongodb.com/pricing )…

At some point our cross-AZ traffic for Elasticsearch replication at AWS was more expensive than what we'd pay to host the whole cluster replicated across multiple baremetal Hetzner servers.

Could we have done better with more sensible configs? Was it silly to cluster ES cross-AZ? Maybe. Point is that if you don't police every single detail of your platform at AWS/GCP and the like, their made-up charges will bleed your startup and grease their stock price.

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#104

Post by travisgriggs »

chubot wrote: post #87
runako wrote: post #59

Earlier quoted context omitted.

> many people just chuck something on us-east-1 and go Even dropping something on a single EC2 node in us-east-1 (or at Google Cloud) is going to be more reliable over time than a single dedicated machine elsewhere. This is because they run with a layer that will e.g. live migrate your running apps in case of hardware failures. The failure modes of dedicated are quite different than those of the modern hyperscaler cl…

It's not an apples-to-apples comparison, because EC2 and Google Cloud have ephemeral disk - persistent disk is an add-on, which is implemented with a complex and frequently changing distributed storage system On the other hand, a Hetzner machine I just rented came with Linux software RAID enabled (md devices in the kernel) --- I'm not aware of any comparisons, but I'd like to see see some It's not straightforward, an…

> by virtue of being more complex

I just wish there was a way to underscore this more and more. Complex systems fail in complex ways. Sadly, for many programmers, the thrill or ego boost that comes with solving/managing complex problems lets us believe complex is better than simple.

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#105

Post by buster »

kosherhurricane wrote: post #22
arbol wrote: post #8

Earlier quoted context omitted.

Atlas wasn't giving us any support for $3K per month. Hetzner at least have some channel to contact them, which is an improvement. That said, if their uptime is rubbish them we'll probably migrate again. Moving back to Atlas is not an option as we were getting hammered by the data transfer costs and this was only going to increase due to our architecture. Thanks for reading!

500GB isn't a lot of data, and $3K/month seems like an extortion for that little data. Having said that, MongoDB pricing page promises 99.995% uptime, which is outstanding, and would probably be hard to beat that doing it oneself, even after adding redundancy. But maybe you don't need that much uptime for your particular use case.

That's all fine and such, but i suppose the SLAs aren't covering your revenue loss.

In fact after looking at https://www.mongodb.com/legal/sla/atlas/data-federation#:~:t... it makes me wonder how much worth the SLA is. 10% Service Credit after all the limitations?

Atlas can take their 10% Service Credit, i wouldn't care. Save the money and chose a stable provider.

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#106

Post by bmcahren »

MongoDB Atlas was around 500% more expensive than in-house every time I evaluated it (at almost every scale they offer as well).

They also leaned too heavily on sharding as a universal solution to scaling as opposed to leveraging the minimal cost of terabytes of RAM. The p99 latency increase, risk of major re-sharding downtime, increased restore times, and increased operational complexity weren't worth it for ~1 TB datasets.

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#107

Post by canucktrash669 »

canucktrash669 wrote: post #103
0xbadcafebee wrote: post #17

> Here's how we managed to cut our costs by 90% You could cut your MongoDB costs by 100% by not using it ;) > without sacrificing performance or reliability. You're using a single server in a single datacenter. MongoDB Atlas is deployed to VMs on 2-3 AZs. You don't have close to the same reliability. (I'm also curious why their M40 instance costs $1000, when the Pricing Calculator ( https://www.mongodb.com/pricing )…

At some point our cross-AZ traffic for Elasticsearch replication at AWS was more expensive than what we'd pay to host the whole cluster replicated across multiple baremetal Hetzner servers. Could we have done better with more sensible configs? Was it silly to cluster ES cross-AZ? Maybe. Point is that if you don't police every single detail of your platform at AWS/GCP and the like, their made-up charges will bleed you…

turns out cross AZ is recommended for ES. perhaps our data team was rewritting the indices too often. but it was an internal requirement. so I think the data schema could have been more efficient to append deltas instead of reindexing all. but none of that will inflate your bill significantly at Hetzner. of course it will at AWS as that's how they incentivise clients to optimize and reduce their impact. and that's how you cut your runway by 3-6 months in compute heavy startups

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#109

Post by mbesto »

kachapopopow wrote: post #6

Always consider if 12 hours of lost revenue is worth the savings. Recently hetzner has been flakey with minimum or no response for support or even status updates that anything was wrong. My favorite was them blaming an issue on my side just to have a maintenance status update the day after about congestion.

If I understand correctly, the author's company provides a CAPTCHA alternative, which presumably means that if their service goes down, all of their customer's logins, forms, etc. either become inoperable or don't provide the security the company is promising by using their service.

This makes me want to use the company's service less because now I know they can't survive an outage in a consistent and resilient way.

Re: We cut our Mongo DB costs by 90% by moving to Hetzner

#110

Post by lxe »

0xbadcafebee wrote: post #17

> Here's how we managed to cut our costs by 90% You could cut your MongoDB costs by 100% by not using it ;) > without sacrificing performance or reliability. You're using a single server in a single datacenter. MongoDB Atlas is deployed to VMs on 2-3 AZs. You don't have close to the same reliability. (I'm also curious why their M40 instance costs $1000, when the Pricing Calculator ( https://www.mongodb.com/pricing )…

I think you underestimate how reduction in complexity can increase reliability. becoming a sysadmin for a single inexpensive server instance carries almost the same operational burden as operating an unavoidably very complicated cluster using a cloud provider.
Post reply on HN