What Is Proof of History? Solana's Cryptographic Clock, Explained

Proof of History (PoH) is Solana's cryptographic clock: a verifiable record that time passed and events happened in a specific order, without validators talking to each other to agree on timing. It is not a consensus mechanism by itself: it feeds Solana's Tower BFT consensus, which is why the network is so fast.

That second sentence is the one most explainers get wrong. PoH is often described as "Solana's consensus mechanism," and that framing leaves readers believing the clock is what secures the network. It is not. PoH only orders and timestamps events; agreement is a separate job, handled by a separate protocol. If you are new to the network itself, start with our guide to what Solana is and come back.

There is also an honest twist, and this guide covers it up front. PoH was the breakthrough that defined Solana's first era. It is now being phased out by an upgrade called Alpenglow. Both halves of that story are below.

The one-line idea: a clock the whole network can trust

Every blockchain has to decide the order transactions happened in, and most networks burn time letting validators message each other until they agree. Solana skips that negotiation. Proof of History gives every validator a shared, tamper-evident clock they can check for themselves, so the order is already settled.

The oldest proof-of-order trick is a photo with today's newspaper in it. The paper proves the photo could not have been taken earlier than this morning, and nobody needs to trust the photographer to accept that. Now extend the trick. Take a photo, print it, and make sure the print is visible inside the next photo you take. Then do it again, and again. Anyone flipping through the stack can verify the order for themselves, because each photo contains the one before it. You cannot reshuffle the stack, and you cannot slip a fake photo into the middle after the fact.

PoH is that stack of photos. It is built with math instead of film, and it grows at high speed, one entry after another. Validators can look at any event in the stream and know what came before it and what came after, without asking anyone (source: Solana Foundation, "Proof of History"). That is what makes it a clock the whole network can trust. The proof of order lives in the record itself, so no trusted timekeeper is needed.

One boundary matters, though. The photo stack proves order; it does not prove the events in the photos are valid. Deciding which version of the chain to accept is a different job entirely. That split, a clock rather than a decision-maker, is where most explainers go wrong. It deserves its own section.

Is Proof of History a consensus mechanism?

No. Proof of History is not a consensus mechanism, and calling it one is the single most common error in Solana explainers. PoH orders and timestamps events. A separate protocol called Tower BFT does the agreeing and decides which version of the chain counts. PoH feeds consensus; it does not replace it.

The confusion is easy to forgive. PoH is Solana's headline invention, and "consensus" is the word people reach for when they describe what makes any blockchain tick. Some widely read explainers even put "consensus mechanism" in their titles when covering PoH. But the two jobs are different, and keeping them apart is the fastest way to understand Solana (source: Helius, "Proof of History, Proof of Stake, Proof of Work — Explained").

Tower BFT is Solana's actual consensus protocol. It is a voting system, adapted from a design family called practical Byzantine fault tolerance. Validators vote on blocks, and those votes stack up until a block is final. PoH casts no votes and cannot decide anything on its own. What it does is hand Tower BFT a timeline so reliable that the voting gets much easier.

A useful way to hold the two apart: PoH is the court stenographer, writing down exactly what was said and when. Tower BFT is the judge who still has to rule. The written record decides no case, but no ruling could be trusted without it. So if PoH is not doing the agreeing, what is it actually doing under the hood?

How Proof of History actually works

Proof of History runs one SHA-256 hash calculation after another, and each new hash uses the previous hash's output as its input. No step can start before the last one finishes. That means the chain cannot be precomputed or parallelized, so a long chain is proof that real time passed.

Start with the ingredient. SHA-256 is a hash function: a piece of math that takes any input and scrambles it into a fixed-length fingerprint. The same input always gives the same fingerprint. Change one character and the fingerprint changes completely. Bitcoin uses this same function for mining; Solana uses it for timekeeping.

Here is the trick. Solana computes a hash, takes the output, and feeds it straight back in as the next input. Then it repeats that, over and over, without pause. Each step needs the previous step's answer before it can begin, so there is no shortcut. More computers will not help, and neither will a cleverer algorithm. Producing entry number 500,000 means a machine truly performed 500,000 steps, one at a time. It is the computer version of counting to a million out loud. Nobody can claim to have done it without spending the time, and here the sequence itself is the receipt (source: Anatoly Yakovenko, "Proof of History: A Clock for Blockchain").

Computer scientists call this kind of construction a verifiable delay function. That means a calculation that is deliberately impossible to rush, so finishing it proves that time passed.

Transactions get their timestamps by being mixed into the stream. When a transaction arrives, its data is folded into the next hash. That pins it to one exact place in the sequence: after everything before it, before everything after it. And while producing the sequence is strictly one step at a time, checking it is not. Verifiers can split a finished chain across many processors and confirm it quickly. Slow to make, fast to check. That imbalance is also the secret behind Solana's speed.

Why Proof of History makes Solana fast

Solana is fast because validators do not have to talk to each other to agree on when transactions happened. The Proof of History stream has already stamped and ordered them. That saved chatter is time the network spends processing instead, which lets Solana batch, pipeline, and run transactions in parallel.

Picture three validators: one in Tokyo, one in Frankfurt, one in New York. Two transactions hit the network a few milliseconds apart. Network lag being what it is, Tokyo sees transaction A first while New York sees B first. On most blockchains, that tiny mismatch forces a debate. Validators swap messages, in rounds, until they settle on one order. The debate repeats for every batch, and while it runs, the network is not processing anything new. The waiting is the cost.

On Solana the argument never starts. Both transactions were already folded into the PoH stream by the block's leader, so each carries a provable place in the sequence. All three machines read the same record and reach the same order without sending a single message about it.

Removing the debate changes what the network can do with its time. The leader can stream transactions without pause. Validators can check one batch while running another, and unrelated transactions can run side by side. The result is the high throughput and sub-second block times Solana is known for. Fees stay low too, because block space is less scarce. For where the clock sits among leaders, validators, and the rest of the stack, see how Solana's architecture fits together. And to see what the design means in practice against the biggest smart-contract network, Solana versus Ethereum walks the comparison.

Speed is the payoff, but the stamped record still has to be made final, and PoH cannot do that alone. That is Tower BFT's job.

How Proof of History works with Tower BFT

Proof of History and Tower BFT split the work. The PoH stream acts as the clock that orders events, while Tower BFT is the voting system where validators commit to one version of the chain. Because the clock is already trusted, validators can vote in fewer rounds, and finality arrives sooner.

Here is one block's journey through both layers:

1. The leader folds incoming transactions into the PoH stream, stamping their order.
2. Validators receive the block and check the stream, confirming the claimed order is real.
3. Validators vote on the block through Tower BFT.
4. Votes stack up, and once enough of the network's stake has committed, the block is final.

The clock makes the voting cheaper than it would otherwise be. In older voting designs, validators have to agree on timing as well as content. That multiplies the messages flying around. Tower BFT skips that step. Every validator's timeouts and vote deadlines are measured against the same PoH stream, so nobody argues about what time it is. Fewer open questions per round means fewer rounds, and fewer rounds means faster finality.

One point rounds out the picture: who gets to be leader, and how much each vote counts, comes from staked SOL. Solana is a proof-of-stake network in that respect, and PoH does not change it. The clock decides when things happened; the stake-weighted vote decides what the chain accepts.

People often ask how all of this compares with Proof of Work and Proof of Stake. The honest answer is that PoH is a different kind of thing entirely.

How is Proof of History different from Proof of Work and Proof of Stake?

Proof of History belongs to a different category than Proof of Work and Proof of Stake. PoW and PoS decide who gets to add blocks and why the network should trust them. PoH only proves when things happened. Solana uses PoH and PoS together, side by side.

Proof of Work and Proof of Stake answer a security question: who can the network trust to write history, and what does it cost to cheat? PoH answers a logistics question: what order did things happen in, and can everyone check it? Because the questions are different, "PoH versus PoS" is not really a contest. Every blockchain needs something in the consensus row. The ordering row is a speed upgrade most chains simply do without. Solana runs PoS for the first job and PoH for the second. That is why "a proof-of-stake chain with a cryptographic clock" describes Solana far better than calling PoH its consensus mechanism.

The mechanics of mining, staking, and their trade-offs are their own subject. Our guides to Proof of Work vs Proof of Stake and what Proof of Work is cover them properly, so this article stays with the clock.

There is one more honest thing to know about that clock: Solana is preparing to retire it.

Is Proof of History going away? Alpenglow, explained

Yes, eventually. An upgrade called Alpenglow is designed to retire Proof of History as the network's clock. It is now in testing on a community validator cluster, targeting Solana's mainnet in late 2026. Blocks would instead be timed by local validator timeouts, with finality expected in roughly 150 milliseconds.

Alpenglow is a ground-up redesign of Solana's consensus layer, written up as proposal SIMD-0326. It has two parts. Votor is a new voting protocol that replaces Tower BFT. Rotor is a new way of spreading block data across the network, though it is slated to arrive later under its own separate proposal, with the current data-spreading system staying in place at first. The most striking choice is the one this article cares about: Alpenglow removes PoH as the network's clock, and validators rely on their own local timeouts instead (source: SIMD-0326, the Alpenglow proposal).

Why retire the breakthrough? Because the clock-plus-voting design, for all its speed, carries overhead. Validator votes travel through the network as transactions and eat capacity. Full finality still takes seconds to arrive. Alpenglow compresses agreement into well under a second by simplifying how validators commit (source: Helius, "Alpenglow: Solana's Great Consensus Rewrite"). As of mid-2026 the upgrade is in testing. It has been running on a community validator cluster since May 2026, with mainnet targeted for late 2026 (source: CoinDesk, "The biggest consensus overhaul in Solana history is officially live for testing"). Until then, PoH remains the clock the live network runs on. Timelines like this can slip, so treat "late 2026" as a target, not a promise.

Alpenglow is not the only sign that Solana's first era is giving way to a second. Firedancer, an independent validator client built by Jump Crypto, has been live on mainnet since December 2025, and a meaningful share of validators now run it (source: MEXC Learn, "Solana Firedancer Explained: Mainnet Launch, 1M TPS Target, and What Comes Next"). It matters here for a simple reason: new clients and a new consensus design are both signs of a network rebuilding its foundations while running at full speed. Our guide to the new Firedancer client covers it fully.

From Blofin's vantage as an exchange running live SOL markets, the practical thing PoH buys our users is ordering before consensus. The chain settles the sequence of events before validators even finish agreeing, and that is part of why SOL deposits and withdrawals confirm as quickly as they do. As Solana moves toward Alpenglow, the shift we are tracking is not the headline number about faster blocks. It is how replacing the PoH clock with local timeouts changes confirmation behavior on the products we support.

So what should a beginner actually walk away with?

What to take away from Proof of History

The takeaway is that Proof of History solved time before agreement. It gave Solana a verifiable clock, so validators could stop negotiating the order of events and start processing them. That one design choice explains Solana's speed, and it matters even as Alpenglow prepares to replace the clock itself.

If you keep three things from this guide, keep these:

1. PoH is a clock. It proves that time passed and that events happened in a specific order, and anyone can check the record without trusting whoever produced it.
2. The clock works because its chain of SHA-256 hashes cannot be rushed or faked. Agreement is still a separate job, done by stake-weighted voting.
3. The design is evolving. Alpenglow, now in testing, is expected to retire the clock in favor of local timeouts, with mainnet targeted for late 2026.

Even after Alpenglow ships, understanding PoH pays off. It explains why Solana was built to be fast in the first place. And the network's willingness to replace its own founding invention tells you how it intends to stay that way.

Frequently asked questions

Is Proof of History the same as Proof of Stake?

No, and they are not rivals either. Proof of Stake is the layer that decides who takes part. Validators lock up SOL, and their influence over block production and voting scales with that stake. Proof of History is the timekeeping layer that orders events for everyone. A network needs something like PoS to choose who it can trust, while PoH exists to make their work faster. On Solana the two run together, one deciding who, the other recording when.

Did Solana invent Proof of History?

Yes. Anatoly Yakovenko described Proof of History in a November 2017 whitepaper, followed by a 2018 essay, before the Solana network existed. He co-founded Solana Labs to build a blockchain around the idea. The network's mainnet beta launched in March 2020. PoH is original to Solana rather than borrowed from an earlier chain, which is a big part of why the idea is so strongly tied to this one network.

What is a Verifiable Delay Function, in one sentence?

A verifiable delay function is a calculation that takes a guaranteed minimum amount of time to produce but only a moment to check, so the finished result is itself proof that time passed. Proof of History is built on this idea. The sequential SHA-256 hash chain cannot be rushed no matter how much hardware you own, yet any validator can check the finished sequence quickly.

Is Proof of History why Solana has had outages?

Not directly. Solana's past outages mostly traced to other parts of the stack: transaction floods that overwhelmed validators, bugs in the validator software, and resources running out under spam load. The Proof of History clock was not the failing part. The network's last full outage was in February 2024, and its reliability record since then has been strong. For the incident-by-incident record and what changed after each one, see Solana's outage history.

Will the Alpenglow upgrade really remove Proof of History?

Yes, that is the stated design. Under SIMD-0326, Alpenglow replaces the Proof of History clock with local validator timeouts, and replaces Tower BFT voting with the Votor protocol. As of mid-2026 it is in testing on a community validator cluster. Mainnet is targeted for late 2026, and that target can move. Until then, Proof of History keeps running exactly as described in this guide, so nothing changes for users yet.

Is Proof of History used by any other blockchain?

Not in any significant way. Proof of History remains effectively unique to Solana among major layer-1 networks. Some Solana-derived projects and networks built on the Solana Virtual Machine inherit pieces of its design. Other fast chains solve ordering with their own techniques instead. Verifiable delay functions appear elsewhere in cryptography research, but embedding one into a blockchain as its clock is Solana's signature.

Researched and written by the Blofin Academy editorial team with AI-assisted drafting. Reviewed by Sabrina Chua (Blofin Academy editor). Primary sources include the Solana Foundation's Proof of History explainer, Anatoly Yakovenko's November 2017 whitepaper and 2018 essay "Proof of History: A Clock for Blockchain," the SIMD-0326 Alpenglow proposal in the Solana Improvement Documents repository, CoinDesk and MEXC Learn reporting on the Alpenglow test cluster and the Firedancer mainnet launch, and Helius Research's analyses of Solana consensus. All facts independently verified against cited documentation current as of July 2026.

This article is for informational purposes only and does not constitute financial advice, investment guidance, or a recommendation to buy, sell, or hold any digital asset. Cryptocurrency markets involve significant risk and you should conduct your own research and consult qualified professionals before making investment decisions. Blofin Academy content reflects the state of public information at time of publication; protocol parameters, fees, and ecosystem data change frequently.