At 19:46 UTC, one Sui address produced 466 of the minute’s 1,127 eligible transactions—41.35% of the total. Its activity was unusually focused: 253 transactions used a DeepBook cancellation workflow and 213 used a limit-order placement workflow. Across the 1,121 transactions with an asserted behavior, those two activities accounted for 555 rows, or 49.51%.
The repeated shape resembles automated order maintenance: the address’s median gap between its own transactions was 0 ms, its 90th-percentile gap was 446 ms, and its longest gap was 1,535 ms. Of its 466 transactions, 425 used the owner-proof pattern and 41 reused one proof across bounded placement runs. The ledger records gas movement for these rows, totaling 415,983,048 MIST, but does not assign application-asset movement to the sender in them. That repetition is consistent with automated execution; it does not establish who controlled the address.
This was not a minute-wide failure wave. Only 9 of 1,127 eligible transactions failed—0.80%, versus 8.36% across the frozen corpus. The nine failures stopped at six different sites, with no site accounting for more than two. One of the concentrated address’s cancellation attempts, Edc7UgvL, stopped at big_vector::leaf_remove; reaching the cancellation workflow does not prove that its application effects persisted. Sui rolls back those effects when the transaction aborts.
The minute’s sharpest movement came from elsewhere. Transaction 728hVojZ1G3m9iRYLJwQn7Sw9knSgKfEa8XaNHqkBJjP split an asset source and passed a resulting coin to a transfer step. Its sender-boundary record shows an outward movement of 40,776.72459463 SUI, roughly 10 times the corpus’s per-transaction 99th-percentile SUI holding movement. This is a boundary movement, not evidence of profit or a priced transfer; the record does not establish the recipient’s identity or whether custody ultimately changed hands.
Other activity remained interleaved through the minute: 193 price-update workflows, 144 campaign-activity records, and 19 swap-route executions were assigned behaviors. For example, FvCZh7vzrpUNSJuwX9KA5GJNX7qJUkAEPkoiFtdNBg5J ran an owner-proof cancellation workflow for an identified NS/USDC order, binding a 0.0115 USDC-per-NS resting price and 4,364.4 NS of resting size—but not proving that the order traded. GmJpmHNMaJ1iJUwGiVfGzk5xaUcXRm7xcQeJ3KykxYKp ran a route touching MAGMA and USDC; its boundary movements were −16.767485505 MAGMA and +5.01119 USDC, but the stored meaning explicitly leaves the exchange outcome unresolved.
The minute’s story is therefore a dense, fast order-maintenance stream with one conspicuous, separately sourced SUI transfer—not a broad system breakdown and not proof of a single operator or profitable strategy.
Further paths: inspect 0xcde6dbe01902be1f200ff03dbbd149e586847be8cee15235f82750d9b06c0e04, the sui/cancel-order/deepbook-owner-proof/v1 pattern, or the outlier transaction 728hVojZ1G3m9iRYLJwQn7Sw9knSgKfEa8XaNHqkBJjP.