[cakoose]

The article describes solutions that use Redis so that multiple app servers can share rate limits. The article's second solution is basically what you're describing, except adapted to work with Redis.

Also, what do you mean by "fixed" memory? Sure, the memory doesn't grow over time, but neither does the memory of the other solutions. Of the solutions listed in the article, this is the most memory-hungry.

[sulam]

The other fixed memory solution that is specifically mentioned suffers from a defect that allows you to go up to 2X past the rate limit by sending 1X on both sides of an aggregation boundary. I thought readers might appreciate an alternative that is also fixed memory but that doesn't suffer from this defect. And sure, the fixed value is larger than other solutions (probably not larger than the Redis solution) but it's likely optimal if you care about being sure the rate limit is never exceeded within your given time interval. YMMV, I am making no claims about universal applicability.

Finally, yes, it's not Redis -- but it could be exposed as a service if you wanted that pretty easily. Operational complexity will exist regardless and depending on your organization different solutions will be appealing for different reasons.