Thread by @buildonbase

Fusaka is coming to Ethereum and bringing more blobs for @base users with it.

Time for a technical deep dive 🧵

We'll go over why this work is important and what Base has been doing to accelerate Fusaka’s timeline in collaboration with @OPLabsPBC, @testinprod_io, and @soneium

First, some background:

Blob TXs were introduced by EIP-4844 in the Dencun upgrade and brought data availability (DA) fees paid by rollup users down by >100x

May’s Pectra upgrade doubled the blob target/block from 3 to 6, which opened up much needed throughput for rollups

Fusaka will allow Ethereum to ramp up to a target of up to 48 blobs/block on a scheduled and safe basis via PeerDAS and BPO hardforks

Why PeerDAS?

Rollups have proven to be an effective way to scale Ethereum and onboard many additional users. Ethereum L1 only provides 96 KiB/s DA at current blob limit, which equates to ~1000 tx/s across all rollups.

PeerDAS helps scale Ethereum by allowing beacon nodes to sample a subset of rollup blob data (instead of all of it), which allows the L1 to handle more blob throughput overall.

PeerDAS will allow blob throughput to scale by ~8x, while keeping user fees low!

As Ethereum client devs were focused on successfully landing Pectra, @base engineers started a working group with @OPLabsPBC, @testinprod_io, and @soneium to accelerate the implementation of PeerDAS in Fusaka.

The @base team identified critical PeerDAS blockers, authored specs for key performance improvements (cell proof computation), and contributed to @lodestar_eth and @prylabs clients to implement features such as column fetching and validator data custody to significantly speed up timelines.

The @OPLabsPBC team’s PeerDAS blog series has provided the ecosystem with deep dive overviews of PeerDAS’ mechanics and goals - breaking down how it works, what it aims to solve, and what we’ve learned from @testinprod_io’s devnet testing.

Check out their work:
blog.oplabs.co/peerdas-and-a-…
blog.oplabs.co/peerdas-devnet…

@testinprod_io has run consistent devnets and published corresponding reports that instill confidence in eventually scaling to 48 blobs/block target and beyond. You can read them here: testinprod.notion.site/Sunnyside-Devn…

These devnets test each client with varied configurations, and the reports pinpoint bottlenecks and drive targeted improvements across client teams.

Their work also included contributing code to and expanding test coverage in @sigp_io’s Lighthouse client.

The @soneium team worked with the @Teku_Consensys client team to analyze the latency in DataColumnSidecar verification runtimes, code review all possible verification paths to avoid double verification, and profile the memory consumption with PeerDAS.

This working group is just one example of how rollups and other L2 teams can play a larger role in contributing to the L1 and achieve shared progress for the Ethereum community.

If you’re interested in contributing to future working groups, reach out to @0xFrancisLi!

Our commitment to scaling extends beyond Base’s own needs and is centered around the long term success of the entire Ethereum ecosystem.

We believe the path to a global onchain economy is paved through collaboration and invite more teams and devs to join our efforts in accelerating Fusaka and future upgrades.

Special thank you to @TimBeiko, @ralexstokes, @corcoranwill, @ethPandaOps, @parithosh_j, @BarnabasBusa, @samcmAU, @_pk910_, @manunalepa, @philngo_, @jimmygchen, @EthDreamer, @fradamt, @nero_eth, @adietrichs, @taem_park, @mininny_, @SunnysidedJ, @jersteth, @protolambda, @sammcingvale, @tyneslol, @EliHaims, @sanjsmehta, and many more.

Their generous onboarding, feedback, and technical advice transformed every milestone into a shared win, and we’re deeply grateful for their partnership.