Solana block capability rose from 60 million to 100 million compute items on July 29, a 66.7% improve that offers unrelated transactions extra room with out altering the separate cap governing its busiest accounts.
Compute items, or CUs, measure computational work carried out by transactions. The block restrict caps the mixture CUs a frontrunner can embrace earlier than the block is full.
The on-chain characteristic account identifies SIMD-0286 as lively at slot 435,888,000, in the beginning of epoch 1009, whereas the activation block timestamps that slot to July 29. The change raised the block restrict by 66.7% mathematically, which the Solana Basis presents as 66%.
The Solana Basis mentioned 11.2% of blocks produced between the July 22, 2025 activation of the 60 million-CU restrict and the newest improve used not less than 56 million CUs.
Why Solana block capability nonetheless leaves a hot-account bottleneck
SIMD-0286 modified solely the general block restrict. The utmost compute that may write to any single account inside a block stays 12 million CUs, whereas the utmost block accounts-data measurement delta stays 100MB. On this context, a sizzling account is shared writable state that many transactions attempt to replace, not essentially one dealer’s pockets.
The capability improve creates extra room for parallel exercise. Transactions touching unrelated accounts can use the extra 40 million CUs, permitting a busy account to crowd out much less of the remainder of the block. Below the previous ceiling, one account’s 12 million-CU allowance represented 20% of the block. It now represents 12%.
The identical math doesn’t give a congested market extra capability. If many trades want to put in writing to the identical account, they nonetheless compete inside the 12 million-CU allowance. A state-heavy workload also can encounter the separate 100MB data-delta restrict even when block compute stays obtainable.
The Solana block capability improve doesn’t straight resolve payment stress. Solana’s payment documentation says an optionally available prioritization payment relies on a transaction’s requested CU restrict and CU worth and raises its probability of being scheduled. The upper block ceiling alone doesn’t present that precedence charges fell for transactions focusing on the identical sizzling state.
Extra capability additionally doesn’t equal measured efficiency. The technical proposal warns that bigger blocks can take longer to execute, sluggish community progress and validator catch-up, and should trigger unexpected points for infrastructure past validators. These are design dangers, not studies of a present incident.
One Aug. 2 block recorded 25,373,012 consumed CUs, 1,328 processed transactions, and 1,069 profitable transactions. One block can’t set up network-wide outcomes or a before-and-after change.
The official improve supplies supply no post-activation before-and-after measurements for transaction inclusion, charges, block propagation, or replay. The improve subsequently establishes the next theoretical ceiling for mixture work. It doesn’t, by itself, show 66% extra noticed throughput or get rid of the native bottleneck that merchants encounter when demand converges on the identical account.
