[plandis]

Another good strategy for load balancing/sharding that always strikes me as simple but also devilishly cleaver is random pick two: https://brooker.co.za/blog/2012/01/17/two-random.html

[amelius]

It looks at mean queue time, not worst case time.

[plandis]

The article linked does, yes. The paper the article is based on (linked in the article) has a proof for worst case load if you’re interested in the details.

Edit: Link to paper http://www.eecs.harvard.edu/~michaelm/postscripts/handbook20...