Protocol Labs Research
About
People
Research
Outreach
Blog
2022-04-20 / Blog
The first graduating class of ConsensusLab projects

When ConsensusLab launched last year, we published a roadmap covering our first 18 months. We are still working towards that same roadmap, with minor adjustments over time to accommodate the changing landscape and externalities.

As we close Q1, we’re pleased to ship the first batch of completed projects, all coming to an end around this time.

🎄 B1: Hierarchical Consensus MVP

We kicked off the B1 project at the end of 2021 with the aim of building an MVP of Hierarchical Consensus (HC). This architectural evolution aims to overcome the throughput challenges of blockchain consensus by horizontally scaling the network. We are proud to have just completed this first project in the HC workstream and now have:

  • An MVP implementation of HC integrated into Eudico (our fork of Lotus), with which you can run your own network with HC! You can spawn your subnets with their own consensus protocols, send messages within a subnet and other subnets, and even run executions using state hosted in multiple subnets with atomicity guarantees.
  • A paper with a description of the architecture of the HC protocol (submitted to DINPS’22)
  • Eudico Garden, a set of basic scripts to spawn your deployment of Eudico in AWS
  • And lots of talks and demos!

⚡ B2 (Pikachu): Checkpointing state onto Bitcoin

Project Pikachu aims to secure any blockchain based on a reusable resource (such as proof-of-space or proof-of-stake) by checkpointing it to the Bitcoin blockchain, thus leveraging the stronger security guarantees afforded by proof-of-work. Our solution leverages the recent Taproot update in Bitcoin and uses threshold multi-signature. We produced:

🌿 Y5: Tendermint BFT as a subnet consensus PoC

Y5 is a project about using a BFT-style consensus protocol in the “Consensus as a Service” model in a Filecoin HC subnet. Tendermint Core was chosen as that protocol and is the first practical consensus implementation inside HC. In this service model, an HC subnet running Tendermint consensus uses parallel Tendermint nodes that return ordered Filecoin messages.

What’s next?

We have several other ongoing lines of research, including the recently started project Y3, which will look at implementing Mir in Eudico, enabling its use in subnet consensus.

If you want to know more about ConsensusLab’s work, visit our GitHub repository, where you’ll find our roadmap, projects, and discussions, or come talks to us in #consensus, in the Filecoin Slack workspace. And remember, we’re hiring!