On “The Power of Choices in Load Balancing with Segment Routing”

Today, and at this very moment (assuming that they stick to the agenda, and I did time-zone computations correctly), our own Yoann Desmouceaux (X11), along with Mark Townsley, are in Atlanta Georgia presenting the paper “SRLB: The Power of Choices in Load Balancing with Segment Routing” at the aspirational 37th IEEE International Conference on Distributed Computing Systems.

Network load-balancers generally either do not take application state into account, or do so at the cost of a centralized monitoring system. This paper introduces a load-balancer running exclusively within the IP forwarding plane, i.e. in an application protocol agnostic fashion – yet which still provides application-awareness and makes real-time, decentralized decisions. To that end, IPv6 Segment Routing is used to direct data packets from a new flow through a chain of candidate servers, until one decides to accept the connection, based on its local state. This way, applications themselves naturally decide on how to share incoming connections, while incurring minimal network overhead, and no out-of-band signaling. Tests on different workloads – including realistic workloads such as replaying actual Wikipedia access traffic towards a set of replica Wikipedia instances – show significant performance benefits, in terms of shorter response times, when compared to a traditional random load-balancer.

Now, of course, you have to go read the paper to get all the details – but here’s a teaser all the same: without introducing more state (or, really, more complexity) than merely distributing requests in a “round robin” fashion (RR in the graph below), the approach developed in this paper (SR4, in the graph below) incurs identical response-times with a much lower server load.

This is literally, “simpler – as fast as, or faster – lower overhead” — and no, you don’t have to “pick two”.

As you no doubt would be able to infer, this paper (and, here’s a preprint) is only the tip of the iceberg of some very cool (some of which, IPv6 SR inspired) ideas that we’re exploring – stay tuned for the next chapter on this objectively very exciting topic.