Skip to content
Advertisement
Anza’s CEO outlines plans to double Solana compute capacity and cut storage costs

Anza’s CEO outlines plans to double Solana compute capacity and cut storage costs

Solana Foundation official brand assets (solana.com/branding)

The Solana engineering firm is pitching a faster, cheaper chain built to rival centralized trading venues

Solana wants to stop rationing its own capacity. Anza’s CEO says the network plans to remove rate limits and increase block sizes.

For a blockchain that markets itself as the fast lane of crypto, that is a notable pledge. It also says something about the customer Solana is chasing: traders who currently do their business on centralized platforms.

What Anza is promising

Anza is the engineering firm that spun out of Solana Labs in January 2024. It builds Agave, the validator client that much of the network runs on.

The broader roadmap comes from Brennan Watt, who took over as Anza CEO in January 2026. Watt laid out Anza’s 2027 plans in a presentation at Solana Breakpoint 2026 on November 17.

The numbers he put forward are aggressive. Anza is targeting a doubling of network compute capacity, aiming for ~500 million compute units per second.

Anza also wants to cut the compute units needed for token-account transactions by roughly 80%.

Advertisement

Storage, propagation and ordering

Storage is the other big target. Through a proposal known as SIMD-0437, Anza aims to cut storage costs by ~90%.

Watt also said a further ~7x reduction in storage costs is expected following recent upgrades. More improvements were expected in the weeks after his talk.

Two more pieces sit under the hood. Rotor is planned to replace Turbine, the protocol Solana currently uses to propagate blocks across the network.

Then there is Constellation, which is intended to improve how transactions are ordered within the protocol.

The road so far

None of this is arriving from a standing start. Alpenglow, a major consensus upgrade designed by Anza, activated on Solana’s mainnet on October 21, 2026.

It cut transaction finality from ~12.8 seconds to approximately 150 milliseconds.

Capacity has been climbing too. A July 2026 increase raised the limit to 100M compute units.

That bump fits with ongoing proposals to eliminate compute-unit limits after Alpenglow. The pledge to drop rate limits looks like the next logical step on that path.

What this means for traders and the derivatives push

The common thread is trading. Anza’s stated aim is high-speed, low-cost on-chain trading that can compete with traditional centralized platforms.

Rotor and Constellation are planned, not live. SIMD-0437’s storage savings and the ~500 million compute-unit target are goals for 2027, so the gap between slide deck and mainnet still needs closing.

The practical checkpoints are straightforward. Watch whether compute-unit limits are actually lifted, whether block sizes grow without stability problems, and whether the storage cost cuts land on schedule.

Disclosure: This article was edited by Estefano Gomez. For more information on how we create and review content, see our Editorial Policy.