Technology•

Anza Wants Solana to Process 500 Million Compute Units a Second in 2027 While Token Transfers Use 80% Less

Anza's 2027 plan for Solana targets double the network's compute capacity, an 80% cut in compute for token-account transactions, lower storage rent and a 96-account transaction limit.

Written By :MINRK
Anza Wants Solana to Process 500 Million Compute Units a Second in 2027 While Token Transfers Use 80% Less

Photo by Brecht Corbeel on Unsplash

Mentioned in This Article

SOL
SOL
Solana
$69.18
2.59%

1. The Solana roadmap for 2027 in brief

Anza, the core development firm behind Solana (SOL), used the Solana Breakpoint 2026 conference in London to set out a technical roadmap for 2027 built on two headline goals: doubling the network's compute capacity from 250 million to 500 million compute units per second, and cutting the compute required for token-account transactions by 80%. Chief Executive Brennan Watt presented the plan.

The roadmap also includes on-chain vector processing, a proposal to reduce storage rent by roughly 90% in stages, and an increase in the number of accounts a single transaction can touch, from 64 to 96. Together the measures are aimed at letting the blockchain carry more activity while lowering what each operation costs.

All of these items are engineering targets, not delivered upgrades. They describe where Anza intends to take the network over the coming year, and each will depend on development, testing and adoption by the validators that run Solana.

Solana is a blockchain designed around high transaction volumes and low fees, and its appeal to users and developers rests heavily on how much activity it can absorb without slowing down or becoming expensive. Capacity planning is therefore central to the network's direction. The plan presented in London addresses that question from several angles at once: the total amount of work the chain can perform, the amount of work its most frequent transactions demand, the cost of storing data, and the complexity of what a single transaction can do.

2. What is a compute unit, and why does doubling matter?

A compute unit is Solana's measure of the processing work a transaction requires, and the network's capacity is the amount of that work it can handle in a given period. Solana currently benchmarks at 250 million compute units per second, and Anza's target is 500 million in 2027.

Every action on the chain, from a simple transfer to a complex trade routed through several applications, consumes a certain number of compute units. Simple operations use few and complicated ones use many. Because capacity is finite, the total compute available sets a ceiling on how much activity the network can process at once.

Raising that ceiling to twice its present level would, in theory, allow the chain to handle twice the computational workload. In practice that means more room for transactions during busy periods, when demand for block space is highest and users compete to have their activity included. A higher limit gives applications more headroom before congestion becomes a constraint.

The qualification matters. A benchmark figure describes what the software is engineered to support, and the throughput observed on the live network depends on real conditions, including the mix of transactions and the performance of validators. The 500 million figure is a goal for the system's capacity, and measured results will show how closely the network approaches it.

3. An 80% cut for the network's most common transactions

The second headline goal addresses efficiency instead of raw size. Anza is aiming for an 80% reduction in the compute units consumed by transactions that involve token accounts, which are the most common type of activity on Solana.

Token accounts are the structures that hold a user's balance of a specific token. On Solana, a wallet does not store all of its assets in a single place; each token a user owns is recorded in its own account. Swapping a memecoin, receiving a stablecoin or moving any asset other than SOL itself involves at least one of these accounts.

Because such operations make up so much of the network's traffic, making them cheaper to process has an outsized effect. Each transaction would cost less to execute, and the capacity no longer consumed by routine token movements would become available for other activity. An efficiency gain of this kind acts like added capacity even before the headline limit is raised.

The two goals reinforce each other. If the ceiling doubles and the most frequent operations need only a fifth of the compute they use today, the effective room on the network for token activity grows by considerably more than a factor of two.

For ordinary users, the practical effect would be felt in the kinds of activity that dominate the chain day to day. Trading tokens on decentralized exchanges, sending stablecoins as payments and moving assets between wallets all rely on token accounts. Lowering the processing burden of those actions is aimed squarely at the bulk of what people actually do on Solana, which is why Anza placed it alongside the capacity figure as one of the two main goals of the plan.

4. Efficiency over hardware

Anza's approach leans on making the software do more with the same resources. Reducing the compute needed for token-account transactions by 80% frees capacity without requiring every validator to buy more powerful machines, and the doubling of headline capacity then adds further room on top.

That distinction is important for a network that depends on independent operators. Validators are the computers that process transactions and secure the chain, and their hardware requirements influence how many participants can afford to run one. Scaling by demanding ever larger machines can raise costs for operators, while scaling through efficiency allows existing equipment to carry more work.

The roadmap also includes on-chain vector processing. This is intended to support advanced computational models running directly on the blockchain, extending the kinds of workloads the network can execute beyond conventional financial transactions.

Seen as a whole, the strategy treats capacity as something that can be recovered as well as added. Every compute unit saved on a routine token transfer is a unit that another transaction can use, so trimming the cost of common operations expands what the network can do without any change to the machines that run it. The planned increase in the headline limit then builds on a base that is already being used more economically.

5. Storage rent and larger transactions

Beyond compute, the plan targets the cost of keeping data on the network. A proposal known as SIMD-0437 addresses storage rent, the fee charged for holding data on Solana, with an overall goal of reducing it by roughly 90%.

Rent on Solana is denominated in lamports, the smallest unit of SOL, and is charged per byte of data stored. Every account on the network, including each token account a user opens, requires a deposit proportional to the space it occupies. Lowering the rate reduces the amount of SOL that users and applications must set aside simply to hold tokens or maintain on-chain data.

The reduction is designed to arrive in phases, not all at once. The first step is a partial move toward the 90% target, and later reductions would need to follow before the full cut is achieved.

A further change concerns the structure of transactions themselves. Anza plans to raise the number of accounts that one transaction can touch from 64 to 96. Complex operations, such as trades that pass through several pools or applications in a single step, must reference every account they read or modify. A higher limit gives developers room to build more elaborate transactions that would not fit under the present cap.

Storage costs and account limits receive less attention than throughput figures, but they shape the experience of building on the network. Rent determines how much capital is tied up in simply existing on the chain, a consideration that grows with the number of accounts an application or its users need. The account limit determines how much can be accomplished atomically, meaning in one transaction that either succeeds completely or fails completely. Easing both removes constraints that developers currently have to design around.

6. Alpenglow gives the plan a track record

The 2027 targets follow an upgrade that has already been delivered. Alpenglow, a change to the way the network reaches agreement on transactions, has gone live and cut transaction finality from about 12.8 seconds to a target of 150 milliseconds.

Finality is the point at which a transaction is considered permanent and cannot be reversed. Shortening it from several seconds to a fraction of a second changes how quickly applications can treat a payment or a trade as settled, which matters for uses such as trading and payments where delays carry a cost.

The fact that an upgrade of that scale has already shipped lends weight to the new roadmap. It makes the 2027 plan more credible than a list of aspirations, since the same development organization has recently carried a major change through to activation on the live network.

Alpenglow and the 2027 roadmap also address different dimensions of performance. Finality concerns how fast a transaction becomes irreversible, while compute capacity concerns how many transactions, and how much work within them, the network can take on. A chain can confirm quickly and still run out of room during heavy demand. With the speed of settlement already addressed, the new plan turns to the volume and cost of the activity being settled.

7. What to watch in 2027

Several markers will indicate whether the roadmap is on course. The first is progress on the later rent reductions under SIMD-0437, which are needed to reach the full cut of roughly 90%.

The second is the rollout timeline for the 96-account transaction limit, a change that developers of complex applications will be waiting on before designing transactions around it.

The third, and the most direct test of the headline goal, is measured throughput as the network moves from 250 million toward 500 million compute units per second. Benchmarks and live network data over the course of the year will show how much of the planned capacity has been realized.

Until those milestones are met, the figures remain objectives. Anza has set out a clear direction for Solana in 2027: more capacity, far cheaper token operations and lower storage costs, achieved chiefly through software efficiency. The delivery of each piece will determine how much of that plan users and developers see in practice.

Related Articles

NEWSLETTERS

Don't miss another story.

Subscribe to the MINRK Newsletter today.

Tick the newsletters you want. No selection? You'll get MINRK Headlines.

By signing up, you will receive emails about MINRK products and you agree to our terms of use and privacy policy.

MINRK Daybook

Market analysis for crypto traders and investors.

EVERY WEEKDAY

MINRK for Advisors

Defining crypto, digital assets and the future of finance for financial advisors.

EVERY THURSDAY

MINRK Protocol

Exploring the tech behind crypto one block at a time.

WEEKLY

MINRK Long & Short

A must read for institutions. Insights, news and analysis delivered weekly.

EVERY WEDNESDAY

MINRK Headlines

The biggest crypto news and ideas of the day.

EVERY WEEKDAY

MINRK Policy

Examining the intersection of cryptocurrency and government.

WEEKLY

MINRK Research

Join thousands of readers who rely on MINRK for data-driven insights on the latest digital asset trends.

MONTHLY