DEV Community

Cover image for 5 Cheap Ways to Host Redis
Jonas Scholz
Jonas Scholz Subscriber

Posted on

5 Cheap Ways to Host Redis

Hetzner, Sliplane, Render, Hashmaps (?!?!), Upstash - Picking a hosting provider for your Redis database can be challenging, especially with all the awesome options available. Analysis Paralysis is real đŸ˜”â€đŸ’«. Who wins the race for the cheapest redis provider?

GO!

1. Hetzner

Hetzner is a German Cloud Provider with Locations in Europe and North America, with a wide variety of compute options including ARM, dedicated, and shared servers. Hetzner is loved by developers, with 70% saying that they want to continue using them according to the latest Stackoverflow survey. Hetzner provides incredibly cheap but at the same time basic servers.

Hosting a simple Redis database isn't actually that complicated! (If you dont need HA, autoscaling, and sharding etc.). Check out this great tutorial from Redis

Self-hosting also means that you will not have any trouble with future Redis license changes!

2. Sliplane

What if you could combine the awesome price of Hetzner with the ease of a PaaS like Heroku, Render, or Vercel? Sliplane is a PaaS on top that gives you push-to-deploy, automatic SSL, a free domain, and more for your Docker apps. Connect your GitHub account and get started in less than 5 minutes for free with a 48 hour trial. Sliplane lets you host an unlimited number of Docker Apps on your server, making it incredibly cheap if you have a large number of low-traffic apps. For example, hosting a frontend, backend, cron jobs, and Redis will only cost you 7 Euros per month.

Redis Start

Disclaimer: I'm the co-founder đŸ€«

3. Don't use Redis!

I knooow, I know. You came for cheap Redis hosting, but do you actually need Redis? Maybe just a simple in-memory hashmap or a local file is enough? Don't overcomplicate your tech stack! Really consider if the additional operational overhead is worth the performance improvements.

While this doesn't work for everything, and especially not for something that needs to be 100% reliable, sometimes it's worth it to think outside the box to save some bucks đŸ€‘

4. Upstash

Upstash is another hosting provider that is loved by developers and has grown tremendously in the last few years.

Upstash is a bit different to the others because its "serverless". For you this mostly means that you only pay for what you use. Thats great if you have low usage, bad if you have a lot. You also trade operational complexity for additional latency, something that Redis probably shouldnt have. Upstash is a great choice if you don't think you can keep a Redis instance up, or if the base prices of the other solutions are too much!

5. Render

Last but not least, another PaaS provider that I want to mention is Render. While Render might look expensive at first, the "Zero DevOps cloud" really makes up for it by being the simplest solution in this list while also providing a free tier for Redis. In the end, the price that your database costs is not everything, you also need to consider the time you are putting in to keep everything running! Sometimes a $20 database is cheaper than a $5 database if you need to work 10 hours less per month, just keep that in mind :)

Conclusion

I hope you learned something new, and always keep in mind to include the price of your own sanity when checking out prices!

Also, I'd love to know where you are hosting your Docker apps. What features do you love, and which do you dislike? Let's discuss it!

Top comments (5)

Collapse
 
andre_adpc profile image
Andre Du Plessis
  1. Don't use Redis!

I fully agree. NEVER use their product EVER, LOL I've dubbed them RCF ("Redis a Corrupted FOSS")
And as Matéush has stated, the likes of DMOC systems are sometimes required. So pick Valkey from the Linux Foundation instead.

May things go well with your Sliplane project, though.

Collapse
 
code42cate profile image
Jonas Scholz

Yeah, valkey looks like a really good alternative at this point. Many are still somewhat locked in to redis though! Thank you:)

Collapse
 
andre_adpc profile image
Andre Du Plessis • Edited

It seems like that indeed. Not surprising, as Redis has only recently "yanked the rug out". There are a few very interesting twists I discovered in the DMOC arena though.
From MS there's Garnet which seems like a "Monsta-gem, chased by a Dragonfly"...

MS' Garnet is MIT licenced vs BSL for DragonflyDB. Both claim they can be used as drop-in replacements for Redis, (memcacheD in DF's case too) and both point out that Redis is basically "frozen in amber" while they are on the move in an F22 Raptor.

Garnet is a new remote cache-store from Microsoft Research, that is designed to be extremely fast, extensible, and low latency. Garnet is thread-scalable within a single node. It also supports sharded cluster execution, with replication, checkpointing, failover, and transactions. It can operate over main memory as well as tiered storage (such as SSD and Azure Storage)

Dragonfly delivers 25x the performance of legacy software

The power of modern cloud hardware can never be realized with legacy software. Dragonfly is optimized for modern cloud computing, delivering 25x more throughput and 12x lower snapshotting latency when compared to legacy in-memory data stores like Redis, making it easy to deliver the real-time experience your customers expect.

Their page title claims to be the fastest in-memory datastore, but they are not comparing themselves very overtly on their own site with Garnet like they do with many others, though. Wonder why?

MS does and even though Garnet appears to outperform DF, one needs to question the objectivity involved I guess. Regardless, both seem like products in for the long-game, and things can get exciting for caching in general.

Now us small-folk need an elegant, tiny, wasp we can employ as a lightweight, easily-configurable "module" or "plug-in" to manage client-side, hybrid and server-side API caching. Something light and framework-agnostic. Like you mentioned in your "Don't Use Redis" section, not every dev-case needs the big guys to keep the house in order.

Any pointers, perhaps?

Thread Thread
Collapse
 
mateush profile image
Matéush

Don't use Redis isn't always an option. Whenever you use serverless (such as Edge function or lambda) the only way to cache data is to use external memory database.

Good luck with Sliplane!

PS. I read your article live on my YouTube channel - youtu.be/L1-agUf4raE

Collapse
 
code42cate profile image
Jonas Scholz

Yes, especially with serverless this might not be an option! Nice video :)

Some comments may only be visible to logged-in visitors. Sign in to view all comments.