Well, you folks now made your business super coupled with AWS. I just have 1 word: Oracle
How we built a serverless architecture with AWS
11–20 of 68 posts
Re: How we built a serverless architecture with AWS
#12Oh boy, I bet that's costing a new car each month.
Re: How we built a serverless architecture with AWS
#13This is not going to scale. Lambdas are hella slow. The cold starts will kill you.
With constant traffic cold starts should not be happening. Also lambdas will stick around for 45-60 minutes before going cold.
Re: How we built a serverless architecture with AWS
#14Oh boy, I bet that's costing a new car each month.
It all depends on how much data they're processing. It looks to be mostly a pay-per-use model.
I'd say their big cost is Kinesis and potentially API Gateway. Lambda is great for this kind of workload (mostly).
Re: How we built a serverless architecture with AWS
#15This is not going to scale. Lambdas are hella slow. The cold starts will kill you.
The only two places where the cold starts will hurt is the API key auth and the JWT auth/posting to kinesis. Plus if it's being called with any kind of decent frequency it won't matter.
Re: How we built a serverless architecture with AWS
#16Oh boy, I bet that's costing a new car each month.
It all depends on how much data they're processing. It looks to be mostly a pay-per-use model. I'd say their big cost is Kinesis and potentially API Gateway. Lambda is great for this kind of workload (mostly).
Top 3 are EC2, DynamoDB and Lambda.
Re: How we built a serverless architecture with AWS
#17This is not going to scale. Lambdas are hella slow. The cold starts will kill you.
With constant traffic cold starts should not be happening. Also lambdas will stick around for 45-60 minutes before going cold.
That's exactly right. Background location tracking leads to constant traffic. Not running into cold starts as an issue.
Re: How we built a serverless architecture with AWS
#18Earlier quoted context omitted.
It all depends on how much data they're processing. It looks to be mostly a pay-per-use model. I'd say their big cost is Kinesis and potentially API Gateway. Lambda is great for this kind of workload (mostly).
Top 3 are EC2, DynamoDB and Lambda.
I don't see EC2 in your architecture diagram. Where are you using EC2?
How are you structuring your dynamo tables? Is there one table that is used much more than another?
Re: How we built a serverless architecture with AWS
#19Earlier quoted context omitted.
Top 3 are EC2, DynamoDB and Lambda.
I don't see EC2 in your architecture diagram. Where are you using EC2? How are you structuring your dynamo tables? Is there one table that is used much more than another?
Not using EC2 directly, though AWS breaks that down cost with a line item for EC2.
Many tables in DynamoDB. Two out of those are most used (equally).
Re: How we built a serverless architecture with AWS
#20Oh boy, I bet that's costing a new car each month.
I used to hate aws for how expensive their bandwidth and storage was, until I started actually using it last year.
I think their new serverless stack is about to leave a lot of devops out of a job.
You can setup a a CI/CD pipeline in about half an hour with amplify, at the previous company I remember it taking a good 3 weeks to get CircleCi up and running properly.
And then moving a microservice over to it is basically 1 command, a few options, mostly just copy over the config from your old express backend with a few changes, and you're done. It's insane.
One other dev I've showed the lighthouse scores of the react stack I deployed on it even said "this should be illegal". And they're right, it's pretty much automated devops, the whole ap now loads in 300ms. If you have server side rendering in your app the static content will automatically be cached on their CDNs.
And if you want to save a bit of money you can just use google firebase for your authentication and db. GraphQl is surpsingly a breeze too as a middle layer if you want to leave your java or .net backend apis untouched.
At the end of the day, nodejs is completely insecure by design, your infrastructure will never be as secure as running it on gcp or aws. That's why you go serverless and stop messing with security and front end scalability.
If they solve the cold-start issue of databases on aurora they will completely dominate the market even more than they already have.