Additional correction: S3 is consistent after new object puts only if you haven't already asked for that object. Otherwise you may get a stale "no object" response.
You can run Python WSGI apps (including Django) using Zappa [1] with a serverless infrastructure.
Do you tell Heroku how many servers you need and how big they should be? Are you in charge of making sure you have the right number, or that the triggers for scaling up / down are working? If so, it is not serverless.
Oh man I was looking for something like this. I have a personal project I wanted to make actually look professional but as a freelancer it's tough to justify big bucks for something like that. But this might work!
AWS has their own v4 signature method that I always thought was neat.
Key benefits:
* Secret not included in request
* Verifies integrity of the message (since its contents are signed)
* Protection against replay attacks
It's probably overkill in a lot of situations, but I've always liked how even if TLS were compromised, all the attacker would gain is the ability to see the requests--not modify them or forge new ones.
I haven't used JWT before, but reading one of the links below, it looks like it covers a lot of the same stuff (although you'd have to implement your own replay protection if you want that).
The issue I ran into is you can't move dollars between exchanges easily, so to rebalance, you've got to switch to something else. So you're profiting from volatility on one side and then betting on it not happening when you rebalance... and for the rebalancing, you have to wait a lot longer (as compared to trading instantly between currencies on the same exchange).
The other issue is that you have to anchor to something, and you can't anchor to the USD generally because even dollars in various exchanges aren't effectively worth the same amount (since you can't get them out as quickly / easily / risklessly in all exchanges). So you've basically got to anchor to US dollars on a certain exchange, and make trades based on the value of, for example, "Litecoin if sold on GDAX." That adds complexity, but might just be the reality of arbitrage.
That and I'm not sure the order books were large enough to make anything other than small amounts of money. The "last sell" price is just that--the price it last sold for. It doesn't take a whole lot of money before you've driven the price toward parity (and your arbitrage opportunity away).
Anyway--for me it was just a winter vacation project more for the experience than expecting real success. And it may actually work better on a week that wasn't last week (which was a bit uncommon). I guess we'll see.
It looks really great. The only concern I have (relative to other serverless features) is that it does have a top-end architectural limit (256 ACUs).
It's probably not worth fretting over, but it is worth mentioning when comparing to DynamoDB. That means in building a serverless web app, you might look at starting with Serverless Aurora, graduating to regular Aurora (as usage increases due to pricing), and then perhaps moving off workloads better suited to DynamoDB as you discover them.
