Why Kubernetes Is the New Application Server • Hacker News Board

Re: Why Kubernetes Is the New Application Server

#51

Post by barrkel » Thu, Jul 12, 2018, 7:36 PM UTC

fauigerzigerk wrote: post #38

Here's my prediction: DevOps is dead. In 5 to 10 years developers will no longer be mentioned in articles like this. There will be specialisation once again, because it makes absolutely no sense to make people whose main job it is to think about use cases, business logic and user interfaces to also deal with stuff like this.

K8s establishes a pattern for implementing many ops tasks as controllers rather than scripts or configs. The level of abstraction will be raised, and more and more of the custom hacks specific to individual companies will become modules you deploy on your cluster.

In this way, devs will be doing ops, but not as ops for any given prod environment, rather encoding operational tasks in reusable software.

Ops themselves will bifurcate, one towards configuring prebuilt controllers, others towards troubleshooting problems in production. The middle ground will hollow out.

Re: Why Kubernetes Is the New Application Server

#52

Post by humbleMouse » Thu, Jul 12, 2018, 7:37 PM UTC

hodgesrm wrote: post #49

humbleMouse wrote: post #47

I think what a lot of people forget is that this is being driven by costs. Companies don't want to pay for disaster recovery, and it's cheaper to set up a k8's platform. The platform itself performs disaster recovery, so now that's one less cost for the company. It's all about saving money on DR. When these companies realize that developers can't handle doing ops AND complex business logic, maybe they'll rethink it.…

Can you define what you mean by disaster recovery? To me it means the ability to recover business applications from a site failure. Assuming you have a similar definition how does Kubernetes solve that problem?

What I mean by DR is that if a data center gets nuked, you still have a replica of your k8's platform running in your other data center(s).

So k8's basically runs a replica of your whole system behind the scenes so if a physical location goes down you still have your system running.

Re: Why Kubernetes Is the New Application Server

#53

Post by markbnj » Thu, Jul 12, 2018, 7:37 PM UTC

fauigerzigerk wrote: post #43

anothergoogler wrote: post #40

Earlier quoted context omitted.

What does "DevOps is dead" mean?

That devs will no longer do ops.

It's more like experienced devs doing ops in support of devs. As with most terms like this ymmv.

Re: Why Kubernetes Is the New Application Server

#54

Post by josteink » Thu, Jul 12, 2018, 7:38 PM UTC

optimuspaul wrote: post #50

fauigerzigerk wrote: post #43

Earlier quoted context omitted.

That devs will no longer do ops.

wait, is that what devops means to you? That devs are doing ops?

That is the predominant definition, so yes I would assume just that.

It’s more or less a logical following of infrastructure as code, because it’s the developers who write and push those files.

Re: Why Kubernetes Is the New Application Server

#55

Post by rb808 » Thu, Jul 12, 2018, 7:38 PM UTC

yobrobroprogr4m wrote: post #17

At least 50% of the neckbeard kvetching about containers is because they don't want to let go of their lovingly hand-crafted servers and mountains of bash scripts and config management crap that they've built over the past 20 years. edit: btw, I'm one of those (former) neckbeards...

Agreed. When people say k8s/docker is complicated, yes it is, but its simpler and more reliable than the shedload of custom build, config & deployment scripts I've seen in past systems.

Re: Why Kubernetes Is the New Application Server

#56

Post by markbnj » Thu, Jul 12, 2018, 7:41 PM UTC

yobrobroprogr4m wrote: post #17

At least 50% of the neckbeard kvetching about containers is because they don't want to let go of their lovingly hand-crafted servers and mountains of bash scripts and config management crap that they've built over the past 20 years. edit: btw, I'm one of those (former) neckbeards...

I'm a neckbeard, literally, and it's even gray :). I'm not sure the implied age gap proves out... maybe. I mean there are always early, mainstream and late adopters of a thing. I remember teaching C++ back in the 90's and there was always at least one person in the back of the room who at some point would say "I can do all this with structs and pointers!" or something to that effect. But yeah, ultimately if you get to be my age and are still doing engineering in this business then it's either because you're still doing it at the same place or you're not afraid of change.

Re: Why Kubernetes Is the New Application Server

#57

Post by zzzcpan » Thu, Jul 12, 2018, 7:42 PM UTC

bsenftner wrote: post #7

taeric wrote: post #4

It is hard not to build this impression of containers. Worse, it seems that the docker image format is massive compared to what most war files were like. This is annoying not just in terms of raw bytes to move around, but in taking stock of what is getting moved around. My team originally pitched how docker solved much of the dependency upgrade management by having layers for each major set of dependencies. That igno…

They are the shit tech that will be abandoned with the next fad. They are a huge boondoggle, solely for the immature leads who have not been coding for decades and know better. This industry is filled with stupid fucks that don't know tech history.

It's not that. The tech is not a solution for everyone's problems to begin with. It's a purely commercial tech made in an attempt to compete with Amazon and sell you stuff. It's very lock-iny, expensive to leave and is crippled on purpose, to make you buy services from Google. Barrier of entry for competition is high too, so it will be pricey.

Re: Why Kubernetes Is the New Application Server

#58

Post by pdimitar » Thu, Jul 12, 2018, 7:43 PM UTC

cphoover wrote: post #41

fauigerzigerk wrote: post #38

Here's my prediction: DevOps is dead. In 5 to 10 years developers will no longer be mentioned in articles like this. There will be specialisation once again, because it makes absolutely no sense to make people whose main job it is to think about use cases, business logic and user interfaces to also deal with stuff like this.

Yes it makes no sense for developers to understand the underlying systems that run their applications. /sarcasm

You definitely have to have a good knowledge about the underlying system, no doubt.

But actually shouldering the burden of creating and maintaining the entire runtime -- and orchestrate it -- is a bit much and stretches people thin.

Re: Why Kubernetes Is the New Application Server

#59

Post by technofiend » Thu, Jul 12, 2018, 7:45 PM UTC

yobrobroprogr4m wrote: post #17

At least 50% of the neckbeard kvetching about containers is because they don't want to let go of their lovingly hand-crafted servers and mountains of bash scripts and config management crap that they've built over the past 20 years. edit: btw, I'm one of those (former) neckbeards...

You have some of those floating around who want to retain their little fiefdoms. I'm at the other end of the spectrum because supporting tens of thousands of snowflakes just makes you want to get a blowtorch. As far as I'm concerned for the right workloads kube can't come fast enough.

The only part I object to is when people get mystical and pie eyed about the idea that putting apps, code, what-have-you in the cloud will mystically make it work well and if it fails you just restart the container or pod and all is well. Which is of course absurd. Bad code is bad code, bad queries run poorly no matter the context. You can buy your way out of some performance issues by scaling up, but that only buys you time at best. IMHO there will always be a need for someone to help people understand why things aren't working well and fix it.

Re: Why Kubernetes Is the New Application Server

#60

Post by lainga » Thu, Jul 12, 2018, 7:46 PM UTC

cphoover wrote: post #41

fauigerzigerk wrote: post #38

Here's my prediction: DevOps is dead. In 5 to 10 years developers will no longer be mentioned in articles like this. There will be specialisation once again, because it makes absolutely no sense to make people whose main job it is to think about use cases, business logic and user interfaces to also deal with stuff like this.

Yes it makes no sense for developers to understand the underlying systems that run their applications. /sarcasm

Maybe that's true, but it also makes no sense for me to be interrupted every 15 minutes by someone asking me "I need you to restart your $deployment for something I'm doing" while I'm trying to develop.