[jameshart]

I'm confused.

If someone tells me that an algorithm scales to 1000 RPS on their infrastructure, and we are having a conversation about scalability, I am interested in what I have to do to increase the number of requests per second by some factor.

Unless what you meant by 1000 RPS was 1000 RPS per server, say. In which case you're claiming that your system scales linearly - adding another server adds another 1000RPS.

[marcosdumay]

Yes, my writing wasn't very clear either.

Your infrastructure imposes a cost structure. Any scalable algorithm will give you sublinear performance gains while you add those costs. At some point, the costs explode, and solving more of the problem isn't economical anymore. Thus, any algorithm will have a maximum capacity at your infrastructure. (1000 RPS is quite low tough, it may have been a bad example.)

[jameshart]

Okay, so yes: you're describing something that does not scale.

[marcosdumay]

If your definition of a scalable system is one that can grow linearly indefinitely, well I have some bad news.

[jameshart]

You’ll need to be more specific. Do you believe there is no such thing as something that scales (sub) linearly?

Economies of scale are a thing. Do you think the marginal cost to AWS of adding more storage servers to S3 increases over time? Or does each additional server amortize a little more fixed cost and only increase the support cost logarithmically, meaning it effectively costs less?