Managed X, X as a service, cloud X... Whether it's infrastructure, hosting, API, or however you define it, what are your favorite services in terms of usability, pricing and overall satisfaction?
For further actions, you may consider blocking this person and/or reporting abuse
Top comments (31)
I'll admit it: I love Heroku! It's the Apple of cloud services to me: as long as you're doing things the Heorku way everything is great. I've worked on a number of teams that didn't have amazing ops personnel like we do at DEV, and in those cases not having to worry about infrastructure was a godsend!
Big, big fan. They marketed push to deploy and ephemeral environments before it was that cool and JAMstack was all the rage. And boy, do they have awesome docs..... So, so good.
dev.to/heroku/how-containers-chang... id a great read.
I worked at a company that had its own data centre about 10+ years ago, and they once took 7 months to provision a database for my team.
I'm thankful every day for Heroku.
I'm the only dev where I work, and the only reason I get away with it is because of Rails and Heroku.
Now, if they'd only support HTTP/2...
Digital Ocean!
Vercel by a mile. Render.com is very promising too
Cloudflare is one of most amazing services..
Not sure if you can call them as could service but they are amazing :)
@digitalocean_staff for hosting :)
Google API's are amazing, just when your project goes out of free borders it becomes quite expensive..
That's kinda short answer from me..
Cloudflare was fine, I guess. That’s until their DNS services were down last week for 20 minutes :angryfist:
There was another incident i remember 6 months ago or so.. 30 min down time..
Issue was the same, somebody messed something up in configs.. I guess..
But still :)
I've seen it down many times but we still using it on my company.
Another thing they could do is custom purge cache when asset version changes (regarding filesize or last modification time), i'm tired of custom purging cache every time we deploy to production :'c
It does not check if the file exists till cache life ends, making file hashing useless for this purpose.
Oh and... cloudflare ignores no-cache headers which annoys me a lot
I am a fan of GCP, Cloudflare, and AWS, mostly in that order.
Azure.I have used
Azureto deploy web apps.I have used
AWSto create a VM.Azure has user friendly UX, its easy to understand.
I heard that
AWSis costly thanAzure.Recently in Microsoft Build 2020, they announced new AI supercomputer .
AWS to Azure services comparison is handy to compare as required.
Interesting, I always look into cheaper stuff of course for personal projects. Which services specifically are cheaper on Azure than AWS that you use? I guess the only thing I really have looked at are VM's specifically on the comparison so haven't checked everything but super curious.
As I said i heard its cheaper. You can also Google azure cheaper than AWS. Below is link from official site.
AWS is 5 times more expensive than Azure for Windows Server and SQL Server
azure.microsoft.com/en-us/overview...
AWS is my favorite hands down. Combined with IaC tool, like Terraform, awsis all you need for day to day basics (no matter of your scale).
I also love Terraform Cloud, cause it allows you to manage TF state without much hassle; Heroku cause it is easy to use (lots of limitation associated with PaS though); and ElephantSQL and MongoDB Atlas for DB as a service. Datadog for monitoring.
Im doing amplify now a days
I really like AWS, probably because I was an early adopter when it came out and I got used to working with it and its quirks. The web console could be a lot better, but I think it's okay-ish most of the time. I find the pricing adequate, not too cheap, not too expensive, and overall I'm satisfied with the SLAs.
Tier 2: Heroku (the original) and Netlify.
Tier 1: I’m loving AWS with Amplify and Serverless framework right now, but I’m keen to try other tier 1 services... Azure is top of my list to try!
Also Cloudflare is doing amazing things with its edge functions that seem to make much more sense than Lambda@edge (Using v8 engine instanced instead of a full container)