Benchmark VM scores
Stay organized with collections
Save and categorize content based on your preferences.
This page shows how to get performance benchmarks for a VM machine series.
How Google creates these benchmarks
Performance benchmarks are created with
PerfKitBenchmarker
on Compute Engine VMs. For optimal parallelization,
the benchmarks are run while specifying a number of threads equal to the number
of vCPUs on each machine type. As a standard option in CoreMark, network
sockets are used for parallelization. For each test of a VM,
CoreMark runs five times and the first of those runs is dropped from the
results.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2026-01-22 UTC."],[],[]]