Live data from Hacker News

Architect – A Serverless Framework

blog.reifyworks.com

41–50 of 76 posts

Re: Architect – A Serverless Framework

#42
post #23

Lambda and SNS are super expensive. I'll skip it.

Lambda is not super expensive if you have only small amounts of traffic... it's actually super cheap in that situation.

It's cheap if you don't use it. The anthem of the cloud.

Re: Architect – A Serverless Framework

#43
post #12

Earlier quoted context omitted.

> The ability to have the infrastructure isolated from the raw code being written. (Think more of a grid computing and automatic planning being included) Can you elaborate? It sounds like you're saying that if you write code that's deployed to a serverless infra you end up with 'serverless code'. This hasn't been my (very limited) experience - where the serverless stuff is just a thin layer around your business logic…

I think they’re saying that you can expose your lower-level atomic APIs as serverless functions, and then, separately, build a “routing and aggregation layer” that exposes APIs that people might actually want to use. And that, for a lot of projects, that routing/aggregation can be entirely specified in a declarative manifest format, rather than involving any coding. This neatly cleaves responsibility: your business l…

Thanks for the more in depth explanation.

Re: Architect – A Serverless Framework

#44

Earlier quoted context omitted.

This is just not true. The discussion is about the syntax (.arc vs yaml), not the information it contains.

it's precisely true; but I get your gist. if you enjoy writing that much cloudformation then by all means have a good time!

I don’t think you get my gist about syntax vs information. I can express the same data in a different format, in the same way that JSON and YAML can often be transformed into each other.

You are pulling a straw man by implying that requires much cloudformation, which is completely unrelated. Besides, CF is a good example of how you can use two different syntax to describe the same information.

Re: Architect – A Serverless Framework

#45

My problem with the serverless movement is database connection pools. Every time a function runs, it has to create a new connection to the Postgres database so it can run a query. This is super slow, and dead in the water. If I have to maintain a separate server to have a global connection pool, I might as well just have a server for my API as well.

I agree with this. I would be comfortable moving more code to serverless if this weren't as big an issue. I think having a GraphQL server as an API to the database may be a good solution, since it can handle a variety of queries/mutations and RPC like behavior. But I still run into this issue a lot whenever I make a new serverless function/project.

Re: Architect – A Serverless Framework

#47
post #5

I’m a big fan of code generation. I can usually adapt to new formats or concepts without too much issue. I feel that while it’s not difficult to build up strings dynamically, the harder part can be overwriting and verifying values, and working with logic. That’s my main issue with YAML over JSON. How would you recommend working in this area with this format, or benefits over JSON?

YAML is far easier for humans, and has a few advanced features (node references) that are REALLY nice for any kind of configuration, since you can re-use blocks. Let's say you have to define a bunch of services, with mostly the same environment variables. You can make a "defaults" node, and then a) reference that node in the service descriptions to re-use those env vars, and b) selectively override those defaults in…

That's good to know, thanks for that tip. I've been using secrets with Kubernetes for sharing things currently, but only in small cases.

Re: Architect – A Serverless Framework

#48

Lambda and SNS are super expensive. I'll skip it.

It can be very expensive. I priced out two scenarios recently for an IoT solution. Using Lambda the cost would be $27,000k a month. Using a LB/EC2 solution with reserved instances the price was $60/month. (We also priced out IoT core, and it was $3,000k a month).

Re: Architect – A Serverless Framework

#49
hey nice project. I had my Show HN recently. I've built a tool to also use it in a serverless fashion. Check it out at https://triggerfs.io

It has been a month since then and the feedback was awesome.

I would love to hear your opinion and how you could possibly use it (use case), since you can do a lot with serverless/FaaS etc.

cheers

Re: Architect – A Serverless Framework

#50
post #2

Great! Another file format for me to learn. Why can't everyone use one of the standard file types, like YAML/JSON?

I really wish people wouldn't use JSON for configuration. The lack of comments and shortcomings in formatting abilities make it particularly ill-suited. { "_comment": "this is a comment" } is cringeworthy to. It's great for transport, but horrible for static file config. YAML is good, but not to everyone's taste. I've been using it for a long time (as a ruby developer) but many people I know find it ugly. I think Arc…

I'm a proponent for JSON but agree that the lack of comments are one of the more annoying things I frequently run into.

edit--Especially when it comes to users editing config files. Imagine a new user trying to edit a web server/db/.bashrc file that didn't have comments to help out?

Post reply on HN