Load is not what you should balance: Introducing Prequal

We present PReQuaL (Probing to Reduce Queuing and Latency), a load balancer for distributed multi-tenant systems. PReQuaL is designed to minimize real-time request latency in the presence of heterogeneous server capacities and non-uniform, time-varying antagonist load. To achieve this, PReQuaL actively probes server load and leverages the power of d choices paradigm, extending it with asynchronous and reusable probes. Cutting against received wisdom, PReQuaL does not balance CPU load, but instead selects servers according to estimated latency and active requests-in-flight (RIF). We explore the major design features of PReQuaL on a testbed system and describe our experience using it to balance load within YouTube, where it has been running for more than a year. PReQuaL has dramatically decreased tail latency, error rates, and resource use, enabling YouTube and other production systems at Google to run at much higher utilization.