How Amazon uses chaos engineering to handle 80k requests per second
1–10 of 119 posts
Re: How Amazon uses chaos engineering to handle 80k requests per second
#2I feel like Amazon search is one of the worst products I've ever used. It is a clusterf/ck of paid advertisements and obviously gamed results. I don't care how many requests/sec you get. If the results are horrible, what does it matter?
Re: How Amazon uses chaos engineering to handle 80k requests per second
#3Surprised that 80K/second is called "massive" for Amazon.com's main search feature.
Re: How Amazon uses chaos engineering to handle 80k requests per second
#4I feel like Amazon search is one of the worst products I've ever used. It is a clusterf/ck of paid advertisements and obviously gamed results. I don't care how many requests/sec you get. If the results are horrible, what does it matter?
'chaos' is a great way to describe the results, to be fair.
Re: How Amazon uses chaos engineering to handle 80k requests per second
#5[dead]
Re: How Amazon uses chaos engineering to handle 80k requests per second
#6I feel like Amazon search is one of the worst products I've ever used. It is a clusterf/ck of paid advertisements and obviously gamed results. I don't care how many requests/sec you get. If the results are horrible, what does it matter?
'chaos' is a great way to describe the results, to be fair.
technically they said 'chaos engineering'... which obviously means they use monkeys slapping their hands on keyboards to write the code that returns the results.
Re: How Amazon uses chaos engineering to handle 80k requests per second
#7Surprised that 80K/second is called "massive" for Amazon.com's main search feature.
Did the math, that's about 7 billion searches a day. That doesn't sound like a lot.
Re: How Amazon uses chaos engineering to handle 80k requests per second
#8Surprised that 80K/second is called "massive" for Amazon.com's main search feature.
It dawned on me that in web software, people talk about req/s from two entirely different perspectives and it's borderline fraud:
req/s from localhost to localhost, and req/s from the Internet to any user.
The latter is actually interesting. People saying you can get 10k req/s from Node.js is stupid. You're not actually getting that on say, a single low-end instance over the Internet, which is what most developers are actually going to do.
Instead, you'll get two orders of magnitude fewer requests per second.
What Amazon is talking about here is most likely non-synthetic, real-world 80k requests per second. Which is actually a decent job.