Live data from Hacker News

Ask HN: Have you shipped anything serious with a “serverless” architecture?

news.ycombinator.com

11–20 of 207 posts

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#11
post #6

Earlier quoted context omitted.

Lot's of empty claims. > some serious shortcoming and a lot more work needs to be put into these serverless platforms Not actionable at all > The approach they're taking I don't think will last No reason given > It really needs a redesign/restructure Nothing here too > And maybe they'll improve it enough by then to make it viable What makes it (un)viable?

That's fair. I'm the founder of a company building a serverless platform. I didn't want to write specifics because it may reveal details about our approach.

Isn't this a great opportunity to sell us on your solution, or get our feedback? You don't need to reveal details of your approach, but what are the actual problems you see?

As a company that has built our most recent site completely on serverless, we've become familiar with some of the issues. Though I definitely would not consider it a no-go.

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#12
post #6

Earlier quoted context omitted.

Lot's of empty claims. > some serious shortcoming and a lot more work needs to be put into these serverless platforms Not actionable at all > The approach they're taking I don't think will last No reason given > It really needs a redesign/restructure Nothing here too > And maybe they'll improve it enough by then to make it viable What makes it (un)viable?

That's fair. I'm the founder of a company building a serverless platform. I didn't want to write specifics because it may reveal details about our approach.

Sounds even more fishy

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#13
post #8
post #2

We've shipped something simple and non-mission-critical in production (URL rewriting for ad placements). It has been pretty much set-and-forget. Last anyone had to even look at it was almost 3 years ago, and afaik it's still working (our ad sales team would be complaining loudly if it weren't). For something peripheral like that, it's nice not to have to run servers for it or devote any energy to keeping it running.…

You front production with an API gateway and version your APIs. You turn migrations into a business validation and testing process rather than a technical dependency; using the load balancer / gateway as the control lever. In this way, serverless can actually be WAY better than traditional methods for dealing with frequent changes. You can have many valid endpoints, but only one “production” endpoint that changes bas…

For some infrastructures and processes that may make sense. The problem for us was that when something didn't work as expected, it was very difficult to determine what was going wrong.

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#14
I have spent the last year and a half building a completely serverless production service on Lambda, API Gateway, and DynamoDB (along with the standard auxiliary services like CW, SNS, Route53, S3, CF, X-Ray, etc.). It was a lot of work establishing new patterns for many of the operational aspects, particularly custom CW metrics and A/B deployments with Lambda traffic shifting, but in the end everything is set up nicely and I'm quite pleased with the end result. We're starting to ramp up traffic now by orders of magnitude (with many more to come) and it's soooooo awesome knowing the stack is pretty much bombproof. Another super-nice thing is all internal authentication and networking being controlled by IAM rather than security groups/VPC/traditional networking - that aspect alone eliminates a tremendous number of headaches.

My biggest complaints are probably DynamoDB eventual consistency (unavoidable when using GSIs), occasional CloudFormation funkiness (though no urgent prod issues yet, thankfully), CodeDeploy CW alarm rollback jankiness (which doesn't tell you which alarm triggered a rollback!!), and lambda coldstarts. But none of these are too terribly concerning and I have faith they'll get incrementally better over time, hopefully.

The biggest cautionary tip I have is we run all our Lambdas with the max 3GB memory both for peace-of-mind and because the underlying EC2 instances have significantly faster CPU. We were seeing weird timeouts and latency initially with Another cost concern I should also mention is that we mitigate cold starts by running multiple canaries using scheduled lambdas (in addition to the standard canary role of generating a baseline of metrics and immediately detecting/alarming on end-to-end issues). We are effectively maintaining a constant warm pool which, in theory anyway, greatly decreases the chances customer traffic will hit cold starts. I'm not intimately involved with the financial aspects but I suspect achieving the same effect with EC2 would be significantly cheaper, at least with respect to infrastructure costs. I would guess, though, that the developer time savings achieved by massively reduced ops burden and overall system simplicity are probably comparable to the increased infrastructure cost, and very possibly hugely outweighing it.

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#15
I've written and shipped numerous sites using Zappa for Python which makes deploying on Lambda/API gateway very simple. https://www.storjdash.com is entirely Lambda based (sorry for no real home page....you can read about StorJ at https://storj.io/)

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#18
Are biggest complaint has been the cold start time. I cant say we have enough production data to see how much its effecting clients though.. We dont have enough constant traffic to keep them primed...

We are all in with AWS so the code build, cloudformation and code pipeline all work really well with lambdas.

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#19
My company uses them as a “shim” to ingest custom data sources into a data warehouse. It produces a really elegant separation of concerns, where the fiddly custom part is isolated in the lambda. I wrote up a blog post about how this works: https://fivetran.com/blog/serverless-etl-with-cloud-function...

Re: Ask HN: Have you shipped anything serious with a “serverless” architecture?

#20
I can't name any names for obvious reasons or give you more hints about what industry this company is in but I just did DD on a very impressive outfit that ran their entire company on Google's cloud platform, it held about 500T of data and held up amazingly well under load.

I was super impressed with how they had set this all up and they were extremely well aware of all the limitations and do's and dont's of that particular cloud implementation.

Obviously there is the lock-in problem, if you ever decide to move you have a bit of work ahead, so build some abstraction layers in right from day 1 to avoid hitting all your code if that time should ever roll around.

And cultivate contacts with your cloud vendor.

Post reply on HN