igneum-network/igneum
1
0
Fork 0
Code Issues Pull requests Releases Activity

HiveOS README: the per-card table from the S_p curve

Browse source 
The "Once a Linux prover ships" column and the sentence above it now follow the proving agent's
S_p curve (bench-log "proving v1" on proving-v1) and the app's default at 894022b (proving on at
24 GB or more, off below): 32 GB mines and proves the prototype shard (28.3 GB alone, 30.0 GB
beside the miner, 2.5 GB spare); 24 GB proves the adopted v1 shard of 30,000 pgas (20.4 GB alone,
about 22 GB beside the miner, approximate) from the fee switch at DAA 210,000 and nothing before
it; 16 GB proves only empty shards alone (13.9 GB) and nothing beside the miner (15.7 GB), so in
practice mines only; 12 GB proves nothing on SP1 6.8.1's GPU server; 8 GB mines only. The sources
line stays; "before the v1-shard row" and the v1-budget caveat are gone, the curve measured it.
README only; selftest.sh unchanged and still passing from df6598f.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
igneum-labs 2026-10-05 20:46:03 +00:00
parent df6598f41a
commit 3d504447ce
1 changed files with 8 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

19
packaging/hive/README.md
View file

@ -11,20 +11,17 @@ workers; no `igneum-prove-host`, no `igneum-prove-export`, no SP1 GPU server. So
80% lottery share and nothing from the 20% proving share until a Linux prover build is published. The Ubuntu rig
installer (`packaging/linux/README.md`, branch `rig-install`, commit dd632c1) carries a prover unit that idles in state
`setup` for the same reason; its per-card table is the fuller version of the one below. What each card could do once
the prover ships, from the app's rule (`app/igneum-app/src/provedefault.rs` on `proving-v1`: 20 GB mines and proves on
one card, 16 GB proves only with the miner paused, under 16 GB mines only) and from the proving agent's measurements
of 5 October 2026, before the v1-shard row (`docs/bench-log.md`, "proving v1": jobs `prover-cost-pc2-pv1b`,
`chain-pc2-pv1c`, `memsweep-pc2-pv1`, `memminer-pc2-pv1`, one RTX 5090):
the prover ships, from the app's default (`app/igneum-app/src/provedefault.rs` at 440fd59 on `proving-v1`: proving on
at 24 GB or more, off below) and from the proving agent's S_p curve of 5 October 2026 (`docs/bench-log.md`, "proving
v1": jobs `memsweep-pc2-pv1`, `memminer-pc2-pv1` and the S_p curve, one RTX 5090, SP1 6.8.1's GPU server):
| Card | Once a Linux prover ships | Measured on 5 October 2026 |
|---|---|---|
| 8 GB | mines only | the prover alone peaks at 13.9 GB on an empty shard |
| 12 GB | mines only | the same 13.9 GB does not fit |
| 16 GB | proves only, with the miner paused | mine-and-prove peaked at 15.6 GB on empty shards and 16.75 GB with chained aggregation |
| 24 GB | mines and proves | 16.75 GB leaves 7 GB; the full prototype shard (28.3 GB with the card to itself) does not fit |
| 32 GB | mines and proves | the full prototype shard beside the miner peaked at 30.0 GB |
The v1-budget shard (about 7 M cycles) is unmeasured and may move these lines.
| 8 GB | mines only | the prover's floor is 13.9 GB for an empty shard |
| 12 GB | mines only; proves nothing on this SP1 build | the same 13.9 GB floor |
| 16 GB | in practice mines only: proves empty shards alone, nothing beside the miner | 13.9 GB alone; 15.7 GB beside the miner leaves nothing for the display |
| 24 GB | proves the adopted v1 shard (30,000 pgas) from the fee switch at DAA 210,000, nothing before it | 20.4 GB alone, about 22 GB beside the miner (approximate, not measured on a 24 GB card); the prototype shard at 28.3 GB does not fit |
| 32 GB | mines and proves, prototype shard included | 28.3 GB alone, 30.0 GB beside the miner, 2.5 GB spare |
## Flight Sheet