The documentation is available here. Feel free to reach us if you would like more details in some documentation chapters or have any questions.

Grandine is a optimised and parallelised client. There aren't many published performance comparisons, but a previous research by MigaLabs may give some insight. We run 50,000 Holesky validators on one of our developer's machine.

Grandine is a lightweight client and needs only ~2.5GB of memory on the mainnet. In order to measure the amount of memory that Grandine actually needs one should stress the operating system to release the shared memory. stress-ng is an easy way to do it:

stress-ng --vm-bytes $(awk '/MemAvailable/{printf "%d\n", $2 * 0.9;}' < /proc/meminfo)k --vm-keep -m 1

Rust is needed in order to build Grandine. We recommend to use rustup.

Some system dependencies are needed, the command below should install it on Ubuntu:

apt-get install ca-certificates libssl-dev clang cmake unzip protobuf-compiler libprotobuf-dev libz-dev

Then the build may take a few minutes:

git clone https://github.com/grandinetech/grandine
cd grandine
git submodule update --init dedicated_executor eth2_libp2p
make release

The compiled binary is available at ./target/compact/grandine.

For faster building (larger binary size) use --release instead of --profile compact.

Docker image can build with a simple command:

docker build .

Cross can be used for Grandine cross-compilation.

Cross-compilation command for amd64 architecture:

make release TARGET=x86_64-unknown-linux-gnu

Cross-compilation command for arm64 architecture:

make release TARGET=aarch64-unknown-linux-gnu

Cross-compiled binaries can be used for Docker images.

Docker build command for amd64 architecture:

docker buildx build \
    --file Dockerfile.cross \
    --platform linux/amd64 \
    target/x86_64-unknown-linux-gnu/compact/

Docker build command for arm64 architecture:

docker buildx build \
    --file Dockerfile.cross \
    --platform linux/arm64 \
    target/aarch64-unknown-linux-gnu/compact/

Grandine is built by Grandine core team led by Saulius Grigaitis. We also involve academia in the early stages of new ideas, however, the optimized production code is delivered by the core team.

Grandine is used in production and the audits have been completed by Matter Labs and X41 D-Sec. The client was also tested by Antithesis team. No serious issues were found. However, always secure your keys with the approach you trust (Web3Signer, Vouch, etc.). Use it at your risk.

It's best to reach us via Grandine Discord or Grandine Telegram. Feel free to join!

Funding public goods is hard. We appreciate your donations via Ethereum address 0x93b10cc89A3D3b5bb6bBB04EfFe104873EF002A9.

Grandine focuses on original consensus core implementations, however it uses a lot of crates developed by the community. For example, Grandine uses rust-libp2p based networking libraries developed by the Lighthouse team since the beginning. Lighthouse's eth2_libp2p library was generic back then and we still use a fork of it now. We also used libmdbx-rs bindings library by Akula maintainer and now we use a fork of it maintained by Reth team. So we focus on the original consensus core because it's the unique value Grandine offers for the community, but we also love to use some great crates developed by other client teams and the community. Grandine would not be where it is now without the efforts of the other client teams and the community! Huge thanks to everyone!