Enabling Hyperscale Web Services

Modern hyperscale web service systems introduce trade-offs between performance and numerous features essential for cost- and energy-efficient operation of data centers (e.g., high server utilization, continuous power management, use of commodity hardware and software, etc.). In this talk, I will present two solutions to bridge the performance vs. cost and energy efficiency gap in hyperscale web services (1) a software system that auto-tunes threading models during system run-time to minimize web service tail latency (OSDI 2018) and (2) a system that exploits coarse-grained OS and hardware configuration knobs to tune limited cost-efficient commodity hardware stock keeping units, to better support their assigned service (ISCA 2019).