Skip to main content

A WebAssembly benchmarking suite for precise performance measurement

Project description

WASURE - WebAssembly SUite for Runtime Evaluation

⚠️ This is a work in progress!

WASURE is a command-line toolkit that helps you benchmark WebAssembly runtimes with clarity. It lets you run benchmarks across multiple engines, manage runtime environments, and generate meaningful visualizations and exports for analysis.

🚀 Features

  • Run and manage WebAssembly benchmarks across various runtimes
  • Install, update, and manage runtimes with simple commands
  • Export results to CSV for analysis or create plots for quick visualization
  • Designed for clarity, reproducibility, and extensibility

🛠 Getting Started

  1. Clone the repository:

    git clone https://github.com/r-carissimi/wasure.git
    cd wasure/wasure
    
  2. Install dependencies:

    pip install -r ../requirements.txt
    

📖 How to Use WASURE

WASURE is structured as a command-line tool with modular subcommands to list, run, compare, and visualize WebAssembly benchmarks. Each subcommand has its own options, allowing you to start simple and scale up your experiments as needed.

python3 wasure.py [OPTIONS] COMMAND

Use --log-level DEBUG to troubleshoot issues and --help under any subcommand for more information.

🔍 Listing Available Benchmarks

See what benchmarks are available:

python3 wasure.py benchmarks list

⚙️ Managing Runtimes

Install, update, and manage supported runtimes:

# View available runtimes
python3 wasure.py runtimes available

# Install a runtime
python3 wasure.py runtimes install wasmtime

# Update or remove runtimes
python3 wasure.py runtimes update wasmtime
python3 wasure.py runtimes remove wasmtime

# List installed runtimes and their versions
python3 wasure.py runtimes list
python3 wasure.py runtimes version

🏃 Running Benchmarks

Run benchmarks with your chosen runtimes:

# Run a single benchmark on one runtime
python3 wasure.py run -b helloworld -r wasmtime

# Run multiple benchmarks on multiple runtimes
python3 wasure.py run -b pystone dummy dhrystone/dhrystone10M -r wasmtime wasmedge wasmer --repeat 3

# Run a raw WebAssembly file directly
python3 wasure.py run -b py2wasm/pystone/pystone.wasm -r wasmtime

Useful Flags

  • --repeat N: Repeat each benchmark N times
  • --no-store-output: Don’t save output, just timings
  • --results-folder <path>: Define custom output directory

📊 Visualizing and Exporting Results

Plot or export results with:

# Plot benchmark output
python3 wasure.py plot results/2025-05-06_10-56-21.json

# Export results to CSV
python3 wasure.py export results/2025-05-06_10-56-21.json

✅ Checking Runtimes Support

The check command allows you to verify if specific benchmarks run successfully on selected runtimes. It is particularly useful when combined with the wasm-features or wasi-proposals benchmark groups. These groups enable you to track which runtime has implemented specific features or proposals.

# Check the wasm features support on all runtimes
python3 wasure.py check wasm-features

# Check the wasi proposals implementation on wasmtime and wasmedge
python3 wasure.py check wasi-proposals -r wasmtime wasmedge

💡 Run WASI benchmarks on runtimes that do not support WASI

Refer to the Replay Merger project for instructions on how to run WASI benchmarks on runtimes that do not support WASI.

➕ Adding New Benchmarks

Refer to the Benchmarks Management Documentation for detailed instructions on adding new benchmarks.

⚙️ Adding or Editing a Runtime

Refer to the Runtimes Management Documentation for detailed instructions on adding or editing runtimes.

⚠️ Known Limitations

  • Platform Support: Linux and macOS are supported. Windows is currently not supported.
  • Path Restrictions: Installers relying on npm (e.g., v8, jsc, spidermonkey) may fail if the runtimes path contains spaces.
  • Non-ASCII Characters: JSC (JavaScriptCore) does not support payload paths with non-ASCII characters.

🤝 Contributing

Contributions are welcome! Feel free to open issues or submit pull requests to improve the project. Please ensure your code follows the PEP 8 style guide for Python to maintain consistency across the project.

📜 License

This project is licensed under the GNU General Public License v3.0.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

wasure-0.5.tar.gz (88.2 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

wasure-0.5-py3-none-any.whl (89.8 MB view details)

Uploaded Python 3

File details

Details for the file wasure-0.5.tar.gz.

File metadata

  • Download URL: wasure-0.5.tar.gz
  • Upload date:
  • Size: 88.2 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.2

File hashes

Hashes for wasure-0.5.tar.gz
Algorithm Hash digest
SHA256 84f9ef0af22d51181b0249549eb13546a220cb51180b88d17daa4164d0fec4de
MD5 82444eb9de6df2523cf5b445e4f8c943
BLAKE2b-256 9884fbdeeb0bc9eb3ba514ddc33acfd516fde82c430ccd7044f44054c2c1c606

See more details on using hashes here.

File details

Details for the file wasure-0.5-py3-none-any.whl.

File metadata

  • Download URL: wasure-0.5-py3-none-any.whl
  • Upload date:
  • Size: 89.8 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.2

File hashes

Hashes for wasure-0.5-py3-none-any.whl
Algorithm Hash digest
SHA256 00a3a1a0ae2a5c30fe06ec85f8916f68ad010b628575f82eb1b0ea907e20f966
MD5 3503710367b4df43bed98a311cdf9b22
BLAKE2b-256 a80ee9b250cdf8d3d188ac11bf3649bf2f6018491877629ff017724b1aa8ca7b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page