...

We use the approach of #now, #next, #later used by foursquare, with a slightly different time horizon. Our #now scale is about 3 months, #next about 6 months, and #later is 12+ months.

Now (up to

...

23.

...

1)

Merge & Post-Merge Clean-up

• Reduction of tech-debt post-Merge relating to addition of new features and specs 
• Fix of outstanding, non-critical issues relating to the Merge

Peering Improvements

The goal is to identify "good" peers fast which will allow Besu to sync to the accurate state faster 

• Improve peer discovery speed 
• Experiment with peer reputation scoring system
• Fix defects on peering

Node Operation Improvements 

• Investigate faster scaling to high volume read RPC traffic 

Next (up to 23.1)

"Shanghai" Upgrade

First fork after Mainnet, whatever it is called. Bullets are notional until committed to on All-Core-Devs calls.

• Withdrawals from the beacon chain 

Client enhancements

• Shanghai devnets & Zhejiang testnet
• Staking withdrawals in Besu
• EOF (delayed, feature complete)
• Execution performance
• Bonsai Stability (bug fixes & auto-heal)

Later (up to 23.4 and beyond)

Cancun Upgrade

• EIP-4488 - Sharding 
• Ethereum Object Format support
  • Code Segments
  • Jump Tables
  • New Opcodes

Modularization of Besu

The goal is to shorten release cycle of Besu, increased adoption of Besu leveraged by composability 

• Assess various approaches to modularization and investigate each option 

EIP Implementation

• EIP-4488 (delayed post-merge)

Client enhancements

• Transaction synchronization optimizations, reduces CPU and Memory use for long-lived transaction pools.
• ETH/66 protocol improvements reducing network traffic

Later (up to 23.4 and beyond)

MEV Besu

• The team will be exploring different options for supporting MEV APIs (most likely this will take place on the consensus layer post-Merge)

"Shanghai" Upgrade

First fork after Mainnet, whatever it is called. Bullets are notional until committed to on All-Core-Devs calls.

• Besu modules for Rollups & L2s 
• Modularize existing multi-use-case for better long term support of public and private networks
• (Later) Support for more chains and environments via Besu modules Ethereum Object Format support
• Code Segments
• Jump Tables
New Opcodes

Optimistic Besu

• The team will be exploring different options for supporting use of the Besu EVM for Optimistic Rollups

EVM Performance

• Continuing work to upgrade the overall performance of the EVM library

Sharding

...

Proof of Stake Enhancements

• Packaged Proof of Stake Images for Mainnet with cleaner code and integrated Consensus layer capabilities

Ongoing Protocol Research

• Besu as an Ethereum reference client in Java
• Verkle Tries
• History / State Expiry research