# GoogleCloudPlatform/PerfKitBenchmarker

PerfKit Benchmarker (PKB) contains a set of benchmarks to measure and compare cloud offerings. The benchmarks use default settings to reflect what most users will see. PerfKit Benchmarker is licensed under the Apache 2 license terms. Please make sure to read, understand and agree to the terms of the LICENSE and CONTRIBUTING files before proceeding.

Repository: https://github.com/GoogleCloudPlatform/PerfKitBenchmarker
Canonical: https://ross.abutalabs.com/products/perfkitbenchmarker
Homepage: https://googlecloudplatform.github.io/PerfKitBenchmarker/
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-26T20:07:14+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 8, longevity 100
- inputs: {"age_days": 4430, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2011, forks 561 (observed 2026-08-28T04:06:05.075820+00:00)

## What it is
PerfKit Benchmarker is a Python tool that automates running a canonical set of benchmarks across cloud providers like GCP, AWS, and Azure. It provisions VMs, installs benchmark workloads, and reports metrics such as latency, throughput, and IOPS for fair cross-cloud comparison.

## Use cases
- compare performance of aws vs gcp vs azure vms
- benchmark cloud instance types with consistent defaults
- measure cloud network throughput between providers
- automate end-to-end cloud provisioning and benchmark runs
- measure disk iops and latency on cloud storage
- benchmark on-premises hardware against cloud offerings

## When to choose
- you need standardized, untuned benchmarks for fair multi-cloud comparison
- you want fully automated provisioning, installation, and execution of benchmark suites
- you need metrics like peak performance, latency, throughput, and time-to-provision across providers

## When to avoid
- you need deeply tuned, vendor-specific performance benchmarks
- you only want to benchmark a single local machine without cloud resources
- you cannot accept the licenses of the underlying benchmark tools it wraps

## Facets
- artifact type: cli-tool
- maturity: active
- function: benchmarking, cloud, cli, workflow-automation
- domain: cloud-computing, performance, developer-tools
- platform: python, cli, cloud
- tags: cloud-benchmarks, multi-cloud, performance-measurement, vm-provisioning, google-cloud, devops, linux, macos

## Member repositories
- GoogleCloudPlatform/PerfKitBenchmarker (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:05.075820+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:01:04.203816+00:00, confidence not recorded.
  - readme: https://github.com/GoogleCloudPlatform/PerfKitBenchmarker (fetched 2026-08-28T04:06:05.075820+00:00, sha 45c1b1601506)
  - homepage: https://googlecloudplatform.github.io/PerfKitBenchmarker/ (fetched 2026-08-29T10:41:04.166890+00:00, sha badb48fecc8a)
- Data as of 2026-08-30T08:39:29.467469+00:00.