Skip to content

Comment on How the end-to-end back-pressure mechanism inside Wallaroo works

Comments

Howdy. I'm the author of both parts of this 2-page series on overload mitigation. Let me know if I've overlooked an important technique for handling too-big workloads or if you have questions (e.g., how we've stitched together TCP's sliding window protocol flow control with Wallaroo's internal flow control).

-Scott

I hope we hear something about this in your talk at DataEngConf SF!

I don't recall which of us is planning on attending DataEngConf SF, but I'm asking around now.

EDIT: Ah, it will be Vid Jain, the CEO of Wallaroo Labs.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.