Katran: A high-performance layer 4 load balancer (opens in new tab)

Their papers[1] are also quite good.

[1] - https://research.fb.com/publications/

How are the licenses? React and Zstandard had licenses that drove away a lot of heavy users (basically anyone who had an active legal department) - and were re-licensed without the onerous patent constraints after an apparent exodus of users (the public outcry seemed to make no difference, but the exodus did).

Receive Side Scaling, not the syndication format :)

There's more info at https://code.facebook.com/posts/1906146702752923/open-sourci...

IPVS cannot be dynamic like XDP can, which can be powered by a reactive bpf program. XDP bypasses every kernel function and basically takes over packet processing at the driver level, so the performance tends to be pretty good. Julia Evans wrote a blog post and has some performance numbers at the bottom of her post: https://jvns.ca/blog/2017/04/07/xdp-bpf-tutorial/

Congrats, you won HN humor of the day award.

FYI, Facebook runs Katran in production: https://code.facebook.com/posts/1906146702752923/open-sourci...

This doesn't really replace something like haproxy. In the diagrams, the "real" servers that they are load balancing to are L7 load balancers like HAProxy (using https://github.com/facebook/proxygen most likely)

The problem this solves is is when you are using haproxy but need redundancy or a single server is no longer capable of handling the load and you need to scale out to 2+ servers.

You could run 4 haproxy servers and use round robin DNS to load balance between them, but as they say, this has the problem of:

> compared to DNS it doesn't need to wait for TTL to redirect traffic from failed L7 lb.

Many people likely use anycast/ECMP for this case, but that has the other issue they mention:

> compared to anycast-based solution, L4 lbs are more resilient to the networking related issues (which triggers mass ECMP reshuffle)/ enable better handling of adding/removing of L7 lbs from the pool as well as provide better support for unequal load balancing.

HAProxy and network layer load balancers are very different kind of beasts. HAProxy requires, for example, you to return your traffic via HAProxy which can make HAProxy a bottleneck, requests are usually smaller than responses. HAProxy will also require you to terminate your connections which will make you lose your source IP address if you cannot get it using proxy protocol.

There are also various other benefits, but network layer balancers do have their own share of downsides as well. Like you need to have pretty big control over your infra in order to use them (generally speaking it's not feasible to use them in public clouds), configuration is more complex, they won't do TLS for you etc.