# Solana Switches From TowerBFT to Alpenglow, Targeting 150-Millisecond Finality

On Wednesday, at slot 459,653,000, Solana validators retired the consensus system the network has used since its 2020 launch and moved vote traffic off-chain. Separately, Ethereum researchers published their own design for broadcasting large blocks faster.

- Published: 2026-10-22T05:48:52.369Z
- Canonical: https://polylog.news/crypto/2026-10-22/solana-switches-from-towerbft-to-alpenglow-targeting-150-mil
- Publisher: Polylog (Crypto desk)
- Section: crypto
- Sources: [CoinDesk](https://www.coindesk.com/tech/2026/10/22/solana-completes-alpenglow-upgrade-here-s-what-it-means-for-users), [Ethereum Research](https://ethresear.ch/t/seacast-fast-lean-broadcast-for-ethereum-with-an-executable-spec-and-a-formal-model/26167), [KuCoin News](https://www.kucoin.com/news/flash/solana-activates-alpenglow-cuts-transaction-finality-to-150-milliseconds), [AMBCrypto](https://ambcrypto.com/alpenglow-solana-upgrade-begins-activation-targeting-150ms-transaction-finality/), [Solana](https://solana.com/upgrades/alpenglow)

Solana activated [Alpenglow](https://solana.com/upgrades/alpenglow) on mainnet-beta on Wednesday. It is the largest change to the network's consensus since the chain went live in 2020. Validators stopped using TowerBFT at slot 459,653,000, at about 14:00 UTC, [according to KuCoin's report of the activation](https://www.kucoin.com/news/flash/solana-activates-alpenglow-cuts-transaction-finality-to-150-milliseconds). Anza, the client developer that designed the upgrade, built Alpenglow to replace the voting system that has produced every Solana block so far.

The upgrade has two parts. A new voting protocol called Votor targets finality in about 150 milliseconds. Coverage of the activation cites roughly 12.8 seconds under the old system. A new block-propagation layer called Rotor replaces Turbine, the mechanism that spread blocks across validators. [AMBCrypto reports](https://ambcrypto.com/alpenglow-solana-upgrade-begins-activation-targeting-150ms-transaction-finality/) that the two components change both how the network agrees on blocks and how it distributes them. [CoinDesk notes](https://www.coindesk.com/tech/2026/10/22/solana-completes-alpenglow-upgrade-here-s-what-it-means-for-users) that users keep their existing wallets and that the main visible change is shorter waits for payments and exchange deposits.

The less visible change matters more for the validator business. Under TowerBFT, validators cast votes as on-chain transactions. Those votes took up block space and cost operators fees every day. Alpenglow moves voting off-chain, which [frees block space and lowers validator operating costs](https://www.kucoin.com/news/flash/solana-activates-alpenglow-cuts-transaction-finality-to-150-milliseconds). Coverage also says the design tolerates a larger share of faulty stake than the old system did. That claim comes from the protocol's designers and will be tested only under live adversarial conditions.

Ethereum researchers are working on the same physical constraint from another direction. A [post on Ethereum Research by researcher Csaba Kiraly](https://ethresear.ch/t/seacast-fast-lean-broadcast-for-ethereum-with-an-executable-spec-and-a-formal-model/26167) introduces SEACAST, a broadcast protocol that comes with an executable specification and a formal model. The post states that Ethereum "has more and more large objects to broadcast, and little time for each." It adds that today's payloads are small and are mostly carried by datacenters, which explains much of why mainnet blocks arrive on time. As blobs grow and home builders try to publish blocks, that advantage disappears. Solana's Rotor and Ethereum's SEACAST are separate answers to the same question: how to move more data to more nodes before a deadline without favoring well-connected operators.

The capital involved is concentrated. DeFiLlama data show $5.96 billion of decentralized-finance value locked on Solana, against $50.16 billion on Ethereum and $6.24 billion on Base. Faster finality does not move that capital by itself. It changes the cost of using the chain for payments and exchange settlement, where waiting time is a direct expense.

## What this means

Faster finality shortens the time exchanges, market makers and payment processors must carry unsettled exposure on Solana deposits. That lowers the working capital they tie up and makes Solana more competitive for stablecoin payments. The decentralization effect points two ways. Off-chain votes cut the fixed costs that push small validators out, but a 150-millisecond voting target rewards operators who sit close to one another, which can concentrate stake geographically. The upgrade itself was announced well in advance and was largely expected. What is new is whether it runs without incident. The precedent is Ethereum's switch to proof of stake in September 2022, which went live without a halt but did not change the asset's price direction in the months that followed. The desk's read is that, if the network runs for several weeks without a liveness failure, major exchanges and payment firms will cut their Solana confirmation requirements within months. A halt, a finality fork, or a measurable drop in the number of independent validators would prove that read wrong.

## What to watch

- Whether the large exchanges cut the number of confirmations they require for Solana deposits. A reduction would show that operators trust the new finality guarantee with their own money.
- The validator count and how stake spreads across geographies and hosting providers over the next quarter. These show whether lower vote costs keep small operators in, or whether the latency target pulls stake into a few datacenters.
- Any liveness pause or finality disagreement in the first weeks after activation. Any such pause or disagreement would test the fault-tolerance claims made for the design.
- Whether Ethereum client teams start implementing SEACAST. Implementation would move it from research to the roadmap for larger blobs.