The Serverless Revolution Has Stalled • Hacker News Board

Re: The Serverless Revolution Has Stalled

#191

Post by nullsense » Tue, Oct 13, 2020, 12:24 AM UTC

jsemrau wrote: post #115

whalesalad wrote: post #93

Earlier quoted context omitted.

I felt your pain immediately and decided to write my own mini-framework to accomplish this. What I have now is a loosely coupled, serverless, frontend+backend monorepo that wraps AWS SAM and CloudFormation. At the end of the day it is just a handful of scripts and some foundational conventions. I just (this morning!) started to put together notes and docs for myself on how I can generalize and open source this to mak…

I never got the hand of cloud formation. I suppose it is nice from a visual (drag and drop) point of view, but I couldn't use it in production and moved on to manage my architecture with terraform.

CDK is a massive improvement over raw CloudFormation, and people are starting to move away from Terraform and to things like CDK or Pulumi.

Re: The Serverless Revolution Has Stalled

#192

Post by rubyfan » Tue, Oct 13, 2020, 12:24 AM UTC

Serverless is the new mainframe. In 10 years everyone will be complaining about their AWS being a mess or their lambdas costing too much. There will be consultants selling modernization to get onto whatever the newest thing is. What is old is new again.

Re: The Serverless Revolution Has Stalled

#193

Post by recursive » Tue, Oct 13, 2020, 12:25 AM UTC

qayxc wrote: post #137

Animats wrote: post #72

Serverless computing refers to an architecture in which applications (or parts of applications) run on-demand within execution environments that are typically hosted remotely. Like CGI or FCGI programs! But with YAML! FCGI really is a "serverless" environment. Launches an application when called for. Starts more copies of the application if there are enough requests. Shuts down idle copies when not needed. Starts a f…

But you forgot that always-on server it lives on. The idea behind serverless includes the fact that you don't pay for what you don't use. So if no requests are served, you owe nothing. Your FCGI solution shuts down idle copies, but keeps the server running...

I don't really understand serverless, but if it's running a process in response to a request, then of course there's still a server, because something received that request.

FCGI has a server running. And serverless has a server running, as long as its listening for requests.

Re: The Serverless Revolution Has Stalled

#194

Post by sabellito » Tue, Oct 13, 2020, 12:25 AM UTC

Nextgrid wrote: post #35

pauldickwin wrote: post #10

Earlier quoted context omitted.

Except that people actually used microservices and microservices are actually additive and helpful.

They are helpful in probably 10% of cases where they are indeed the right tool for the job and the benefits outweigh the downsides. For the majority of companies out there it's just a tool to overcomplicate your stack and turn it into an engineering playground so you can justify 2-3x the headcount despite no significant productivity increase. But hey, at least your company can now be giving talks about how they solve…

Although your overall point seem reasonable, "some companies need it, some don't", throwing "10%" and "majority" around doesn't help with credibility.

Re: The Serverless Revolution Has Stalled

#195

Post by somurzakov » Tue, Oct 13, 2020, 12:25 AM UTC

is there an alternative to Serverless using kubernetes? like a service that spins up a pod, checks out code from repo, runs the job and then shuts down the pod.

Re: The Serverless Revolution Has Stalled

#196

Post by wiremine » Tue, Oct 13, 2020, 12:25 AM UTC

I've seen a few serverless projects that are really, really jacked up. 1000s of functions glued together with little intentional design... basically one big pile of spaghetti code.

Things like the Serverless framework[1] help, but ultimately today's serverless architectures feel like they forgot Dijkstra's warning: that the "Go To Statement Considered Harmful." [2]

I'm still big on the approach for the right use cases: variable work loads are a good example.

But ultimately, to be truly useful, we need to see either a) better coupling with non-serverless approaches, or b) a "serverless 2.0" approach that builds in the good parts and fixes the bad parts. (But please don't call it serverless 2.0... those of us who lived through "Web 2.0" will thank you)

[1] https://www.serverless.com/

[2] https://homepages.cwi.nl/~storm/teaching/reader/Dijkstra68.p...

Re: The Serverless Revolution Has Stalled

#197

Post by sabellito » Tue, Oct 13, 2020, 12:26 AM UTC

Aeolun wrote: post #91

Because it sucks . We’ve been working with serverless for 2 years now and we’re still dealing with issues for which the only solution appears to be to move back to ECS.

Care to expand? You seem to be one of the few people on this comment section that has used it in production for a long time.

Re: The Serverless Revolution Has Stalled

#198

Post by abraxas » Tue, Oct 13, 2020, 12:26 AM UTC

Aeolun wrote: post #91

Because it sucks . We’ve been working with serverless for 2 years now and we’re still dealing with issues for which the only solution appears to be to move back to ECS.

Can you be more specific? I'm not being facetious, my company is embarking on this very path and I would like to hear what lies ahead.

Re: The Serverless Revolution Has Stalled

#199

Post by wiremine » Tue, Oct 13, 2020, 12:27 AM UTC

nickjj wrote: post #25

Kind of surprised the article didn't mention lack of reasonable development environment. At least on AWS, the "SAM" experience has been probably the worst development experience I've ever had in ~20 years of web development. It's so slow (iteration speed) and you need to jump through a billion hoops of complexity all over the place. Even dealing with something as simple as loading environment variables for both local…

> It's so slow (iteration speed) and you need to jump through a billion hoops of complexity all over the place. Even dealing with something as simple as loading environment variables for both local and "real" function invokes required way too much effort.

Honestly, it reminds me of PHP development years ago: running it locally sucked, so you need to upload it to the server and test your work. It. Sucked.

Re: The Serverless Revolution Has Stalled

#200

Post by nullsense » Tue, Oct 13, 2020, 12:29 AM UTC

inoop wrote: post #147

nickjj wrote: post #25

Kind of surprised the article didn't mention lack of reasonable development environment. At least on AWS, the "SAM" experience has been probably the worst development experience I've ever had in ~20 years of web development. It's so slow (iteration speed) and you need to jump through a billion hoops of complexity all over the place. Even dealing with something as simple as loading environment variables for both local…

Take a look at the [CDK]( https://aws.amazon.com/cdk/ ) if you haven't already. It lets you define your infrastructure using TypeScript, which then compiles to CloudFormation. You can easily mix infrastructure and Lambda code in the same project if all you're doing is writing some NodeJS glue Lambdas which sounds like what you're looking for. There's a couple of sharp edges still but in general it just 'makes sense'.…

CDK made IaC accessible to me. I hated raw CloudFormation and never bothered with it because of that reason. I had a crack at Terraform, but never got passed the learning curve before my enthusiasm died.

Currently using some CDK in a production app and finally I found a way of doing IaC I actually enjoy.