Original author: Filecoin Network
Filecoin Fast Finality (FIP 0086: Fast Finality in Filecoin (F3) ) is one of the most transformative upgrades to the Filecoin consensus layer since the mainnet launch. After nearly two years of intensive research and development, F3 is now ready for mainnet deployment and is expected to bring breakthrough progress to the entire ecosystem! By reducing the finalization time from 900 blocks to only about 2, F3 brings a full range of substantial benefits:
Faster transaction confirmations: Token holders and Dapp users will see transactions completed in minutes instead of hours.
Enhanced cross-chain experience: Oracles and bridges will complete transactions and state changes faster. Imagine a future Dapp that can seamlessly issue storage transactions from other chains to Filecoin and receive confirmation within minutes of the data being successfully stored on Filecoin!
More efficient lightweight client: Node operators such as RPC providers and storage providers can use lightweight clients to improve accessibility.
Streamlined data storage process for storage providers: F3 simplifies the data storage process for storage providers without the need for additional hardware or operational changes. Storage providers using SP clients such as Lotus Miner, Venus, or Curio or participating in the Ramo network can significantly improve data storage efficiency. By reducing the time to issue storage transactions by more than 2 minutes through F3 , storage providers can achieve faster data storage processes and provide users with optimized SLA performance.
And more!
Changing the consensus protocol of a live network is not an easy task, and to ensure a smooth and safe transition to the mainnet, the change has been rigorously audited and tested in the calibration network. Although F3 was originally upgraded for nv24 , two critical issues were discovered during the later trials, which can be found here . These issues have now been resolved and the fixes are being verified on calibration.
Next step
The F3 team is now in the final stages of preparation testing on the mainnet. This includes deploying the updated protocol under production traffic and following the launch test plan . To provide more visibility, we have also released two F3 monitoring dashboards ( mainnet , calibration testnet ) for the community to track progress, monitor block set finality, node performance, etc. If you want to get the latest testing updates, please subscribe to this forum or join the #fil-fast-finality slack channel.
Notes for Node Operators
As we enter the final passive testing phase on mainnet, it is important to understand F3 ’s bandwidth usage patterns. These patterns are primarily driven by two key factors:
Message volume: F3 needs to hear from participants who control at least 66% of the network power to make decisions.
Message size: The more chunks that need to be processed, the larger the message will be.
👋🔔Expected bandwidth usage during Bootstrap
During the bootstrap phase, F3 must complete all block heights since the last finalization, which is approximately 900. This finalization of the extension chain causes a temporary spike in bandwidth usage that typically lasts 5-10 minutes before settling back to baseline levels.
Other known high-bandwidth scenarios
F3 achieves finality in so-called instances: you can think of an instance as a meeting where everyone shows up to decide something. At least 66% of attendees need to be present for the meeting to adjourn. After F3 bootstrap, each instance typically contains blocksets of a few blocks in normal operation. This is because F3 , as the name implies, is fast finality, and in the vast majority of cases it finalizes blocksets within 2 blocks.
However, in some cases, F3 finalization may lag behind the progress of the chain. For example, a meeting may take a long time to end due to issues such as lack of participation or network connectivity. Because the Filecoin chain is constantly growing, instances that come after longer instances now require a longer chain to complete. Longer chains mean larger messages, because every message exchanged as part of an F3 instance contains information that the chain is finalized. Larger messages will cause a surge in bandwidth usage, which will drop to the baseline once the F3 instance catches up with the chain, that is, completes the block set within the last few block heights.
If you see unexpectedly high bandwidth usage associated with F3 testing for extended periods of time, please collect logs from your implementation, especially from the F3/gpbft and F3/certexchange log levels, and share them in the #fil-fast-finality Slack channel . The F3 team is actively monitoring performance and will work with you to investigate and resolve the issue.
🏁 F3 to start
Over the next 2-4 weeks we will continue to gather data on the soundness of the F3 implementation through passive testing, and our current plan is to launch F3 with nv25 .
While we are still waiting for final data to confirm whether F3 is ready, the F3 team plans to propose an off-cycle upgrade to nv25 to Filecoin implementers and core developers. If no critical issues arise in the final stage of testing, we will make the following recommendations for the release of F3 :
Early December 2024 - Release calibration nv25 upgrade and deploy F3 final code ( F3 is already activated and running)
Mid-December 2024 - User side releases mainnet nv25/F3 activation block height set
January 2025 - Mainnet nv25/F3 activation
We will keep you informed of progress and provide updates to the community if any plans change. Thank you for your support of this exciting upgrade!
You can follow the FilOz blog and Twitter for the latest updates on Finality’s unveiling.
Special thanks to other F3 core implementers: Kubuxu , Masih , Stebalien , annorth , BigLep ; other Lotus implementers rvagg and Forest implementers Hailong and ruseinov for your hard work in bringing F3 to life.
