Microsoft doubles down on Kubernetes for Azure
blogs.dxc.technology
Microsoft doubles down on Kubernetes for Azure
1–10 of 61 posts
Re: Microsoft doubles down on Kubernetes for Azure
#2Edit: Thanks to replies explaining that in fact AKS doesn't charge for the master! I'm leaving the rest of this comment standing because it's an honest reply to the article posted - Microsoft isn't disingenuous, this article is just wrong.
> Be that as it may, Microsoft is offering AKS for free. You will only pay for the virtual machines (VM) that you use for managing your Kubernetes cluster. Microsoft says, “Unlike other cloud providers who charge an hourly rate for the management infrastructure, with AKS you will pay nothing for the management of your Kubernetes cluster, ever. After all, the cloud should be about only paying for what you consume.”
This is disingenuous. We're talking about k8s, so GKE is the comparison point. Yes GKE charges a flat hourly fee for cluster management, regardless of the number or size of the nodes[0]. But GKE doesn't charge for the master instances (which are hidden away behind the GKE product). So AKS will not charge for the cluster management, but will charge for the master instances. Neither pricing model is more "for free" than the other.
Also, if we have to think about the master instance sizes on AKS because we're paying for them, then that's one thing we have to manage on AKS that we don't have to manage on GKE - GKE adds more value by more completely managing our cluster.
[0] pedantic: there's actually no fee for clusters of less than 6 nodes.
Re: Microsoft doubles down on Kubernetes for Azure
#3I really, really hope aws joins the game. re:invent is around the corner.
Re: Microsoft doubles down on Kubernetes for Azure
#4Edit: Thanks to replies explaining that in fact AKS doesn't charge for the master! I'm leaving the rest of this comment standing because it's an honest reply to the article posted - Microsoft isn't disingenuous, this article is just wrong. > Be that as it may, Microsoft is offering AKS for free. You will only pay for the virtual machines (VM) that you use for managing your Kubernetes cluster. Microsoft says, “Unlike…
GKE does charge for the masters for clusters of 6 or more nodes. See https://cloud.google.com/kubernetes-engine/pricing. Microsoft's offering does not seem to charge you for the masters ever, so that does seem like a differentiator.
Re: Microsoft doubles down on Kubernetes for Azure
#5Edit: Thanks to replies explaining that in fact AKS doesn't charge for the master! I'm leaving the rest of this comment standing because it's an honest reply to the article posted - Microsoft isn't disingenuous, this article is just wrong. > Be that as it may, Microsoft is offering AKS for free. You will only pay for the virtual machines (VM) that you use for managing your Kubernetes cluster. Microsoft says, “Unlike…
Statement's like Microsoft's are horrible; "the cloud should be about paying for what you consume" well, Kubernetes consumes at least one master node in order to orchestrate a cluster.
But, Google/GKE does charge for management of the cluster. A constant fee of $109/mo for 6+ nodes. Its not technically the charge for the master node, but that cost to Google + other things is likely where that $109/mo comes from.
Re: Microsoft doubles down on Kubernetes for Azure
#6This is a good thing; Microsoft will increase competition in this space by applying its expertise in dev tools to Kubernetes.
I just hope that MS doesn't add in too many platform-specific pieces that would encourage vendor lock-in.
For example k8s ingress right now is based on controllers and GCE controllers support/don't support a wide variety of things compared to nginx ingress... these areas of Kubernetes make me worried about potential future switching costs.
Re: Microsoft doubles down on Kubernetes for Azure
#7Edit: Thanks to replies explaining that in fact AKS doesn't charge for the master! I'm leaving the rest of this comment standing because it's an honest reply to the article posted - Microsoft isn't disingenuous, this article is just wrong. > Be that as it may, Microsoft is offering AKS for free. You will only pay for the virtual machines (VM) that you use for managing your Kubernetes cluster. Microsoft says, “Unlike…
GKE does charge for the masters for clusters of 6 or more nodes. See https://cloud.google.com/kubernetes-engine/pricing . Microsoft's offering does not seem to charge you for the masters ever, so that does seem like a differentiator.
You write "Microsoft's offering does not seem to charge you for the masters ever". The article says "You will only pay for the virtual machines (VM) that you use for managing your Kubernetes cluster". Do you have a source that disagrees with the article?
Edit: to those replying correcting the article, if you were to create a top-level comment on this thread correcting the article, we could upvote your comments and better correct the misinformation. I have edited my own comment explaining what you have dug up. Thanks for your research!
Re: Microsoft doubles down on Kubernetes for Azure
#8I really, really hope aws joins the game. re:invent is around the corner.
Well, you can use something like kops, although it's a little more difficult to get started, I guess. This new tool from Azure is basically the same thing.
It's a little interesting since they already have ECS, which isn't that bad either, but it seems like everyone just wants Kubernetes. Will they do both, or provide a migration path?
Re: Microsoft doubles down on Kubernetes for Azure
#9This is a good thing; Microsoft will increase competition in this space by applying its expertise in dev tools to Kubernetes. I just hope that MS doesn't add in too many platform-specific pieces that would encourage vendor lock-in. For example k8s ingress right now is based on controllers and GCE controllers support/don't support a wide variety of things compared to nginx ingress... these areas of Kubernetes make me…
> I just hope that MS doesn't add in too many platform-specific pieces that would encourage vendor lock-in.
That's Microsoft's main strategy though. Example: Users can't even install Firefox or Chrome on Windows 10 S.
Microsoft announced: "Apps that browse the web must use the appropriate HTML and JavaScript engines provided by the Windows Platform."
I wouldn't be the slightest bit surprised to see similar things happening to their cloud platform as soon as they are able to do them without users' noticing.
Re: Microsoft doubles down on Kubernetes for Azure
#10This is a good thing; Microsoft will increase competition in this space by applying its expertise in dev tools to Kubernetes. I just hope that MS doesn't add in too many platform-specific pieces that would encourage vendor lock-in. For example k8s ingress right now is based on controllers and GCE controllers support/don't support a wide variety of things compared to nginx ingress... these areas of Kubernetes make me…
Fortunately there's a bunch of ingress controllers you can use: Traefik, Voyager, HAProxy, and probably several others. And it's surprisingly trivial to write your own.
So ingresses is not currently a weak point in relation to vendor lock-in happens. And Kubernetes already supports plenty of non-Google tech; as an open source project, Kubernetes is refreshingly non-Google-focused (there are a bunch of players, notably Red Hat and Microsoft, ensuring this).
As an aside, the current ingresses (including the Nginx one and Google's own GLB one) all have annoying deficiencies (subpar support for TLS and per-route timeout settings among the biggest). Ingress support is, and has been for a long time, Kubernetes' weakest point. For example, GLBs max out at 20 TLS certs (ridiculous if you're hosting many customers on a SaaS solution) and default to a timeout of 30 seconds (you can't control this using ingress annotations; you have to manually go in and edit the backends via API or UI) (which doesnt work for big streaming requests and WebSockets). These are also very trivial problems compared to the complex ones that are being solved by big new features in Kubernetes proper, so it's a bit surprising that ingress implementations are lagging to this extent.