3a54ed89-2
Anza Targets 500M Compute Units Per Second on Solana by End of 2027
Anza's 2027 roadmap doubles Solana compute capacity to 500M CUPS, cuts token-account compute 80% and storage rent 90% via SIMD-0437, with account limits rising to 96.

Outputs
Anza targets 500 million compute units per second by end of 2027, up from a current 250 million benchmark.
Roadmap aims to cut compute needed for token-account transactions by 80%.
SIMD-0437 targets ~90% storage rent reduction; first step lowers cost from 6,960 to 5,080 lamports per byte.
Per-transaction account limit rises from 64 to 96.
Alpenglow upgrade, live October 21, 2026, cut finality from ~12.8 seconds to a 150-millisecond target.
Anza, Solana's core development firm, has set a target of 500 million compute units per second by the end of 2027, doubling the network's current benchmark of 250 million. CEO Brennan Watt presented the roadmap at Solana Breakpoint 2026 in London, pairing the capacity goal with an 80% reduction in the compute units required for token-account transactions.
What does the 2027 roadmap change?
The throughput target is the headline figure. Solana currently benchmarks at 250 million compute units per second; hitting 500 million would in theory let the chain process twice the computational workload.
The second goal targets Solana's most common activity type. Token accounts hold balances of specific tokens on the network — every swap of a token other than SOL touches one. Cutting the compute cost of those operations by 80% has a double effect: each transaction costs less to process, and freed capacity opens up for other network activity.
The roadmap also includes on-chain vector processing, designed to support advanced computational models running directly on the blockchain.
How far do fees fall?
A proposal called SIMD-0437 targets storage rent, the fee for keeping data on the network, with an overall reduction goal of roughly 90%.
The first step cuts the cost from 6,960 to 5,080 lamports per byte. Lamports are the subunits SOL divides into. That initial cut is a partial move; the remaining reductions toward the ~90% target would follow in later steps.
Anza also plans to raise the number of accounts a single transaction can touch from 64 to 96.
What has Solana already shipped?
The roadmap builds on delivered work. The Alpenglow upgrade, implemented on October 21, 2026, cut transaction finality from roughly 12.8 seconds to a target of 150 milliseconds.
That delivery record matters for credibility. Anza's approach leans on efficiency rather than hardware expansion: an 80% compute reduction for token-account work frees capacity without requiring every validator to buy a bigger machine, and the capacity doubling adds room on top.
The caveat is timing. These are targets, not shipped changes, and the main capacity goal sits at the end of 2027.
What happens next?
Watch three markers over the coming year: progress on SIMD-0437's later rent reductions, the rollout timeline for the 96-account transaction limit, and measured throughput as the network moves from 250 million toward 500 million compute units per second.
via Crypto Briefing (Source)
More from Tom Whitfield
Show full bio
News editor covering technology and product news in crypto and web3 industry at Mempool Brief.
1,453 articles