Getting in the Zone for Successful Scalability

TL;DR

This paper explores the universal scalability law (USL) and its relation to queueing models, revealing how different scalability zones affect application performance and guiding tuning for optimal scalability.

Contribution

It demonstrates the equivalence of USL to queueing models and characterizes scalability zones, offering insights for application tuning.

Findings

USL is equivalent to synchronous queueing in a load-dependent model

Scalability zones are defined by USL, Amdahl's law, and Gustafson scaling

Throughput measurements span all three zones, informing tuning strategies

Abstract

The universal scalability law (USL) is an analytic model used to quantify application scaling. It is universal because it subsumes Amdahl's law and Gustafson linearized scaling as special cases. Using simulation, we show: (i) that the USL is equivalent to synchronous queueing in a load-dependent machine repairman model and (ii) how USL, Amdahl's law, and Gustafson scaling can be regarded as boundaries defining three scalability zones. Typical throughput measurements lie across all three zones. Simulation scenarios provide deeper insight into queueing effects and thus provide a clearer indication of which application features should be tuned to get into the optimal performance zone.