Solana Tests Alpenglow: Finality Cut From 12.8 Seconds to 150 Milliseconds

Solana has begun testing the Alpenglow upgrade on a public testnet, a move that could cut transaction finality time from about 12.8 seconds to just 150 milliseconds. Finality is the moment when a transaction is considered irreversible, and it has long been one of the main obstacles to blockchain adoption for everyday payments.

In the current system, crypto exchanges wait for finality before crediting deposits, cross-chain bridges hold funds until a transaction is considered final, and merchants need assurance that a payment cannot be reversed. During that waiting period, users face the risk of price fluctuations and operational uncertainty. Alpenglow promises a major change on this front.

The upgrade replaces the TowerBFT consensus mechanism with a voting protocol called Votor. Instead of accumulating validator votes on-chain over 32 slots, Votor lets validators exchange votes directly and reach agreement in just one or two rounds. Applications continue to run as usual, and users do not need to change wallets or alter how they send funds.

Alpenglow is not an overnight project. For more than four months, the upgrade ran on a small network built specifically for testing. The move to Solana's public testnet is the next test: whether all the computers and services already running the blockchain can switch over simultaneously. The testnet uses tokens with no monetary value, so developers can find issues and restart the network without endangering user funds.

Anza, one of Solana's core development companies, recommended Agave 4.3 to all mainnet validators on September 21. Previously, this version was rolled out gradually to operators controlling 10% and then 25% of the total SOL securing the network. However, two alternative validator programs built by Jump Crypto, Firedancer and Frankendancer, do not yet support Alpenglow. Both are designed to reduce Solana's dependence on a single software client, so Alpenglow's first migration will run entirely through Agave.

"The absence of Firedancer and Frankendancer from this trial means risk is concentrated on Agave. If there is a bug, the impact could be broader," a developer noted in a paraphrased technical document.

For Indonesia, this development has direct implications. Solana has become one of the favorite blockchains for payment applications and stablecoins in the Asia-Pacific region. If 150-millisecond finality becomes reality, financial service providers in Indonesia that are beginning to explore asset tokenization and cross-border payments could consider Solana a more competitive infrastructure. Finality speeds approaching those of traditional payment systems such as BI-FAST or RTGS could accelerate blockchain integration into everyday financial services.

Even so, adoption will not happen automatically. Validators in Indonesia and globally need to upgrade their software, and wallet and exchange ecosystems must adjust their operational processes. In addition, the temporary reliance on Agave raises questions about network resilience if that software experiences disruption.

Looking ahead, the question is not only whether Alpenglow succeeds technically, but whether this increase in finality speed is enough to change market behavior. If Solana can consistently offer transaction settlement below 200 milliseconds, a new standard for layer-1 blockchains will take shape, and competitors such as Ethereum, Avalanche, and rollup-based networks will have to respond. In the long run, users and developers in Indonesia will be the final judges: whether finality speed becomes a decisive factor in choosing a network, or whether other factors such as cost, security, and regulation remain more dominant.