5 ms·
This story resonates with me so much. An avid cloud platforms user in the past, now I just use Dockerfile + fly.io combo. Dockerfile gives you a total freedom w
by garganzol 3y ago
This story resonates with me so much. An avid cloud platforms user in the past, now I just use Dockerfile + fly.io combo. Dockerfile gives you a total freedom without vendor lock-ins and forceful deprecations. I have never been calmer and happier before that switch. Now, only I decide when to upgrade - and usually it is as easy as editing a text file pointing to a new package version.
IMHO, the only thing cloud providers are genuinely useful for is a bottomless data storage. But even that aspect is now isolated by Rclone in all my apps since migrating off the cloud. Apps just talk with the file system, which is then abstracted to a cloud storage of choice by Rclone.
- scarface_74 3y agoYes, I’m sure major corporations are going to trust their entire enterprise to fly.io and all they need is Docker and storage.
- VHRanger 3y agoI mean, you could still avoid vendor lockin by having proper build scripts and deploying directly to EC2 instances and their equivalent. The lockin happens when DevOps gets lazy and start using a bunch of cloudy functions.
- scarface_74 3y agoAnd now you’re going to spend more money on cloud just to treat it like an overpriced collocation center without any of the benefits of moving to it. Have you ever done a large migration at scale? I have - even moving over a lot of VM hosted services involves your project management organization, regression testing, networking, firewall, training, security, working with outside vendors, and the list goes on. The average enterprise uses 254 SaaS services. I have never in 5 years heard any CTO of any company of any size worry about “lock in” when considering their cloud choices. Source: 2 years as an architect responsible for integrating acquired companies before a private equity owned company went public, two years leading the “application modernization” effort at a startup that exited 10x revenue, 3 years working at AWS in the Professional Services department and now working at a smaller consulting shop. And I got Amazoned a couple of months ago. I have no particular loyalty toward Amazon
- secondcoming 3y agoMaybe they don't worry about it because they don't think there's anything to worry about it. The most expensive part of switching cloud providers is data transfer. That's where the lock-in is, Kubernetes vs VMs isn't that big a deal at all; annoying sure. For the record, my employer uses both Kubernetes and VM. The deployment is pretty much the same.
- scarface_74 3y agoThe most expensive part is the staff and man hours. Everyone here is completely ignoring the organization complexities.
- icedchai 3y agoTotally. Migrating anything is a PITA, even if it's simple. If it's complex and involves tons of code changes to eliminate "lock in", or more likely, move from one flavor of lock-in to another, you better think seriously whether it's even worth bothering. Developer time is very, very expensive, and any time you spend "migrating" is time not spent building features that add value. Except in rare cases, the end customer probably doesn't care if your stuff runs on AWS, GCP, or Azure.
- diarrhea 3y agoI guess the point is then: once you’re migrated over, you live happily. For a while. The next migration will come. It’s only a question of when. What then? You’re now so far removed from simple Unix-like principles, tools and platforms that migration will be harder by an order of magnitude (and you noted how much work it is!).
- scarface_74 3y agoAnd what about the 250 other SaaS services that your company uses? Are you going to suggest that your medical software doesn’t integrate with third party EMH/EMR systems? That your call center and sales team doesn’t integrate with Salesforce? What about your dependency on Workday? ServiceNow? Your project management software? All of the Microsoft software? Yes even VMWare. There is a reason that AWS has a VMWare offering.
- garganzol 3y agoWhile you are being sarcastic, all corps, big and small, will be there one day. The situation is like using UTF-8 when everybody still rely on ASCII.
- paulgb 3y agoI take the “H” in IMHO (“in my humble opinion”) to mean “this is true to my situation, which does not mean it’s true for everyone”
- refulgentis 3y agoIdk. I don't think every major corporation can do that
- djha-skin 3y agoOne of our engineering teams wanted to use fly.io instead of kubernetes that I was running in AWS for the company. They really liked its ease of use and they really wanted to use it. Unfortunately fly.io only allowed 5,000 concurrent TCP connections, and there's was a chat app that needed to maintain many more web sockets than that for each instance that was running (More like 50,000). So they ended up switching to kubernetes and AWS. Fly.io is nice, but as far as scale is concerned, it's entry level.
- garganzol 3y agoThe TCP connection limits are in place only when a central HTTP/HTTPS routing/termination is used. But you can turn that off by a config parameter in fly.toml file and by allocating static IPv4/IPv6 addresses for your app. By doing so, your app will be then fully responsible for handling all HTTP/HTTPS traffic by itself without external caps.
- miroljub 3y agoIt's funny how people spend a lot of time and, in the end, choose a worse (unfavourable?) product just because of their inability to find a config switch that would solve all their issues and enable them to use their tool of choice.
- scetron 3y agoThere's quite a few reasons why it happens though. Perhaps the feature was available, perhaps there was no documentation or poor documentation. Perhaps this is more complex in some way? We don't really know.
- ctvo 3y agoThere's a another reason to avoid fly.io outside of platform capabilities: their outages caused by growing pains. If it were me, I'd circle back in a year or two.