tl; Ph.D
- Merge progress — minor spec updates, engineering full steam forward 🚂
- No progress in consumer range. Be egocentric, lead a minority consumer!
Merge updates
Initially — implausible work by all engineering groups on the Kintsugi dash, culminating within the launch of the Kintsugi Merge testnet. Wonderful to see 3 govt shoppers and 5 consensus shoppers for a complete 15 totally different {couples} working on a unified entrance.
Kintsugi🍵, the primary long-term Merge testnet, was not with out pleasure. The #TestingTheMerge the trouble hammered the testnet with transactions, unhealthy blocks, and quite a few different chaotic inputs, creating some errors in state transitions, synchronization, and extra. We anticipate finding such bugs in early testnets, however with every iteration the shoppers develop into extra steady.
Restarting the furnace 🔥🧱
The groups recognized an vital downside just a few weeks in the past. This was a discrepancy in Engine API (how the PoS consensus layer drives the execution layer) semantics associated to how execution layer shoppers truly work in observe. The issue is that, in some contexts, the consensus layer by chance brought about an sudden load on the execution layer.
Engineers then realized that if the engine’s API semantics had been slightly extra versatile, the 2 layers may work extra harmoniously. This led to a delicate however essential modification Engine API and associated breaking apart particular version.
At this time, Range specification🔥🧱 has been launched, and engineers are busy pushing out modifications. On the finish of this dash, groups intention to convey production-ready deployments to a brand new testnet for public use. Preserve your eyes open to take part.
From there, groups will transition from public testnets to proof-of-stake earlier than making ready for the mainnet.
Measures of consumer range
Michael Sproul launched a brand new wave consumer range metrics utilizing its new fingerprint mechanism. Sadly, the consumer distribution of validation nodes has not moved within the final 6 months.
The number of consumer implementations of the consensus layer gives Ethereum and its customers with a singular and sturdy resistance to software program failures and assaults. Customers achieve some resiliency by utilizing a minority consumer no matter community composition, however the community itself good points resiliency towards a number of key validator distribution thresholds.
If one consumer:
- Doesn’t exceed 66.6%, error/bug in a single consumer can’t be finalized
- Doesn’t exceed 50%, error/error within the alternative of 1 consumer’s fork can’t dominate the chain head
- Doesn’t exceed 33.3%, error/bug in a single consumer can’t disturb finitude
Judging by the looks of the fingerprint engine, the Prysm remains to be above the 66.6% mark.
I wish to give a giant shout out to groups, people and communities that take consumer range significantly (Template A, Exhibit B). Operating a minority consumer isn’t solely wholesome for the community, but in addition safer for the person consumer’s funds.
Be egocentric (rational)! Lead a minority consumer 🚀
