Virtual Machine Warmup Blows Hot and Cold

TL;DR

This paper introduces a statistical method to analyze VM warmup behavior, revealing that many VMs do not reliably reach peak performance during warmup, challenging common assumptions.

Contribution

It presents an automated changepoint analysis approach to determine steady state and peak performance in VM benchmarks, exposing variability in warmup outcomes.

Findings

Many VMs do not reach steady state of peak performance

Only up to 43.5% of VM-benchmark pairs consistently reach peak performance

Warmup behavior varies significantly across different machines

Abstract

Virtual Machines (VMs) with Just-In-Time (JIT) compilers are traditionally thought to execute programs in two phases: the initial warmup phase determines which parts of a program would most benefit from dynamic compilation, before JIT compiling those parts into machine code; subsequently the program is said to be at a steady state of peak performance. Measurement methodologies almost always discard data collected during the warmup phase such that reported measurements focus entirely on peak performance. We introduce a fully automated statistical approach, based on changepoint analysis, which allows us to determine if a program has reached a steady state and, if so, whether that represents peak performance or not. Using this, we show that even when run in the most controlled of circumstances, small, deterministic, widely studied microbenchmarks often fail to reach a steady state of peak performance on a variety of common VMs. Repeating our experiment on 3 different machines, we found that at most 43.5% of <VM, benchmark> pairs consistently reach a steady state of peak performance.