[evanj]

Thanks for the link, the abstract seems relevant. I agree: a control algorithm seems like it should work. The challenge is figuring out the metrics and parameters to make something that works "well enough" for most applications, like TCP does for networks. I would really love for someone to figure that out, so we don't run into this very often.

[ot]

Yeah, it's something that needs to be tuned per-service. It is quite easy if the service is X-bound for some local resource X (CPU, disk, flash, network card), but if it is bottlenecked on external service calls it can be quite challenging to define a representative stress metric.