|
|
|
|
@ -24,7 +24,7 @@ Internal adversarial pass, not an independent review. Every sentence in this fil
|
|
|
|
|
| Q2b | Hot set over drawn programs | 32 epochs under the devnet era (epoch seeds `igneum-adv-cache-2/epoch/k`) at 2^23 nonces each (2^30 reads per program); 32 drawn eras queued | the same plant | the same gates per program; the census line PASS/FAIL | All 32 devnet-era and all 32 drawn-era programs done (section 2.3): overall every program clear on the hot-set test (items 0.9993x to 1.0075x of the windowed control at the top 0.1 percent, lines 0.9992x to 1.0006x; quarter and half shares the model to four digits). Four more sites of the Devnet 3 kind over 1.2x (era-fixed-20 site 11 1.26x, era-drawn-2 site 13 1.27x, era-drawn-10 site 7 1.23x, era-drawn-15 site 14 1.51x) and four at 1.04x to 1.16x in 48 x 16 = 768 sites; the same base programs under the devnet era (rotation 29) read 1.00x to 1.01x at those sites: the era rotation decides whether a product's biased low bits land inside the item index (section 2.3): era-fixed-20 site 11 (instr 46, r2, last shadow writer `mulhi`: one item read 512 times at a mean of 4, top 0.1 percent 1.26x, source hi16 bucket at 52x uniform) and era-drawn-2 site 13 (instr 55, r6, last shadow writer `sub` after a chain: chi2/dof 2.04, 1.27x). With Devnet 3's site 0 that is 3 sites in 36 programs (576 sites), 0.5 percent of sites, 8 percent of programs; each worth 0.03 to 0.1 percent of a hash's reads to a store (section 4.1) | FINDING (a class with its mechanism: 10 of 1,056 sites over 1.2x, 19 over 1.04x; half of the drawn-era programs carry one against 2 of 32 under the R = 29 devnet era; sub-class A a product in the source's last writes whose low bits the era rotation leaves inside the item index, 30 of 31 rotations do; sub-class B a warp-uniform source once in 16,000 warps; worst 1.51x, worth 0.05 percent of a hash's reads) and BOUND; drawn-era census complete. Ledger AP-F8-6. The class v5 lane's read (its log `docs/design/class-v5-harness/v5-listed-adv-cache-2-full.log` on branch class-v5): the class v5 0.995 distinct-index floor at the 2^20 acceptance sample refuses 0 of these 61 programs (every v5 draw lands on the v4 attempt); the eight over 1.2x read 0.9965 to 0.9997 at their worst site (era-drawn-28 0.9969, era-drawn-25 0.9994), the 1.04x to 1.2x band 0.9986 to 0.9998, clean ones 0.9999 to 1.0000: the floor sees few-item hot sets (adv-accept's class), not this diffuse excess over the top 0.1 percent of a window |
|
|
|
|
|
| Q3(1) | Steering by the item index t | `adv-cache-2 steer`, 2^20 random t x 32 bit flips, rounds 0, 1, 7, days 20730 and 20733 | `--plant t-low` (round-0 address = t AND mask): identity rows at P = 1.0, dead rows for bits 22..31, 40,960 equal pairs at +231,704 sigma. Fired | every cell within 6 sigma of 0.5; equal pairs within 6 sigma of 8 | Day 20730: worst cell 3.44 sigma; equal pairs 13 / 4 / 6 of 2^25 against 8.00 (+1.77 / -1.41 / -0.71 sigma). Day 20733: worst 3.95 sigma; pairs 8 / 11 / 7. 2 x 2,112 cells, none beyond 4 sigma | PASS (BOUND at 2^20 items x 32 flips per day) |
|
|
|
|
|
| Q3(2) | Steering by the day key: the weak-day scan | `adv-cache-2 days`, 16,384 consecutive chain days x 2^16 items (2^30 derivations), per day the segment and depth histograms and a same-size control | `--plant quarter-lines --plant-day 20733` on 8 days x 2^14 items: day 20733 listed alone at segment chi2_z +1,089.7 and max bucket 22 (+14.1 sigma) against the control's 11; the 7 clean days 10 to 11. Fired (19:23:49Z) | no day with segment chi2_z over 6 or a depth bucket beyond 6 sigma; the per-day max against the control's | 16,384 days (20730 to 37113) x 2^16 items = 2^30 derivations in 707 s on 32 cores; 0 mirror mismatches; worst per-day segment chi2_z +3.96 (day 36139); worst per-day max segment bucket 31 at mean 8 (+8.13 sigma, day 29452) against the control's worst 30 (+7.78); the per-day max-bucket histogram REAL 20:44 21:1797 22:5877 23:5156 24:2335 25:800 26:279 27:66 28:24 29:3 30:2 31:1 against CONTROL 20:30 21:1793 22:5974 23:5034 24:2335 25:825 26:285 27:72 28:29 29:6 30:1; worst depth z_max +4.96; 0 days flagged | PASS (BOUND: no weak day in 16,384, 45 years of chain days) |
|
|
|
|
|
| Q3(3) | Steering by the parameters: the window layer's concentration | `adv-cache-2 windows`, 4,096 programs under the devnet era and 4,096 under drawn eras, the expected share per quarter from the 16 sites' window draws; measured on the real programs in Q2 | `--plant all-quarter` (every site forced to one quarter): top-quarter share 1.0000 on 8 programs. Fired | none: the distribution is the result (by design above uniform) | Real programs: devnet top quarter 0.42187 of reads (model 0.42188), top aligned half 0.53125; Devnet 3 top quarter 0.39063 (model 0.39062), top half 0.71875. Exact distribution from the draw: top quarter mean 0.3382, median 0.3281, p99 0.4688; top half mean 0.5811, p99 0.75. 4,096 drawn programs: mean 0.3377 / 0.5806, max 0.6250 / 0.8750 (section 3.3) | FINDING (designed, counted; priced in 4.1) |
|
|
|
|
|
| Q3(3) | Steering by the parameters: the window layer's concentration | `adv-cache-2 windows`, 4,096 programs under the devnet era and 4,096 under drawn eras (both done), the expected share per quarter from the 16 sites' window draws; measured on the real programs in Q2 | `--plant all-quarter` (every site forced to one quarter): top-quarter share 1.0000 on 8 programs. Fired | none: the distribution is the result (by design above uniform) | Real programs: devnet top quarter 0.42187 of reads (model 0.42188), top aligned half 0.53125; Devnet 3 top quarter 0.39063 (model 0.39062), top half 0.71875. Exact distribution from the draw: top quarter mean 0.3382, median 0.3281, p99 0.4688; top half mean 0.5811, p99 0.75. 4,096 devnet-era programs: mean 0.3377 / 0.5806; 4,096 drawn-era programs under drawn eras: 0.3378 / 0.5808, tail probabilities 0.0283 / 0.0208 against the closed form's 0.0277 / 0.0199 (section 3.3); the era draw does not enter the window shares (they come from the program stream), as the two censuses agree | FINDING (designed, counted; priced in 4.1) |
|
|
|
|
|
| Q4 | The price: ops per item and the SRAM column at the measured hit rates | arithmetic in section 4 from Q1, Q2, Q3(3); the v2 binary prints the stride and hottest-lines costs (nearest held line walk) and the hottest-items store per program | hand points: stride f = 1/2 is 0.5 evaluations per read; f = 1/64 is 31.5; the model's f = 0.25 row is 899,072 ops per hash | a BREAK is a store whose measured hit rate puts ops per hash under the model's row for that f by over 10 percent | the only measured excess over f is the window layer's (an aligned half serves 53 to 78 percent of reads: the model's f = 0.5 row overstates the recompute share by up to 1.8x) and the line reference multiplicity (a hottest-lines half store hits 57.8 percent instead of 50, at a higher miss cost than the stride) | FINDING against the model's table, not the hash; v2 rows RUNNING |
|
|
|
|
|
| GPU | any GPU measurement | none in this lane | | | no GPU on either box | BLOCKED |
|
|
|
|
|
|
|
|
|
|
@ -276,7 +276,7 @@ The one designed non-uniformity. Each load site's window is the dataset, an alig
|
|
|
|
|
| Devnet 3 epoch 0 (measured over 2^31 reads) | 0.39063 | 0.71875 |
|
|
|
|
|
| Exact distribution from the draw (win uniform on {0, 1, 2}, off uniform; 16 sites; enumerated over the 74,613 compositions, scratch `python3`, reproduced in section 3.3 below) | 0.2500 / 0.3281 / 0.3382 / 1.0000 (p90 0.4062, p99 0.4688; P(over 0.45) 2.8 percent) | 0.5625 / 0.5811 / 1.0000 (p90 0.6562, p99 0.7500; P(at least 0.75) 2.0 percent) |
|
|
|
|
|
| 4,096 drawn programs under the devnet era (item 13, 1,708 s wall on 80 threads; `win` draws 0/1/2: 22,047 / 21,856 / 21,633 of 65,536) | 0.2500 / 0.3281 / 0.3377 / 0.6250 (p10 0.2812, p90 0.4062): the closed form to three digits | 0.5625 / 0.5806 / 0.8750 (p90 0.6562) |
|
|
|
|
|
| 4,096 under drawn eras (item 17) | queued behind the weak-day scan and the 2^32 census | |
|
|
|
|
|
| 4,096 drawn programs under 4,096 drawn eras (8 shards of 512 on box 2, 00:04Z to 00:12Z, 46 to 62 s each on 32 cores; `win` draws 0/1/2 within 2 percent of a third each) | 0.2500 / 0.3281 / 0.3378 / 0.6250; P(at least 0.45) 0.0283 (closed form 0.0277) | 0.5625 / 0.5808 / 0.8750; P(at least 0.75) 0.0208 (closed form 0.0199) |
|
|
|
|
|
|
|
|
|
|
The plant `all-quarter` (every site forced to one quarter) printed 1.0000 on every program: fired.
|
|
|
|
|
|
|
|
|
|
@ -367,9 +367,10 @@ The first load of a hash (site 0 in iteration 0) depends only on the nonce, the
|
|
|
|
|
| Q2b era-drawn 21 to 24 (shard s5, 172 s) | 2 | 172 s | 32 | 0.02 |
|
|
|
|
|
| Q2b era-drawn 25 to 32 (shards s6, s7, about 175 s each) | 2 | 350 s | 32 | 0.03 |
|
|
|
|
|
| Q2 devnet 2^25 under v2 (item 430) | 2 | 84 s | 32 | 0.01 |
|
|
|
|
|
| Q3(3) windows drawn 4,096 (8 shards) | 2 | 437 s | 32 | 0.04 |
|
|
|
|
|
| Pod-hours | none (no GPU, no rented pod) | | | 0 |
|
|
|
|
|
|
|
|
|
|
Running total at 00:05Z (8 October): about 2.2 box-hours of run by wall x threads / 96; the lease waits (box 1's line shard s2c has waited since 22:09Z with 0 free adv-class cores) are not counted by wall x threads / 96 (the boxes ran at load 400 to 600 for the first two hours, so the CPU actually consumed is under that). The 8-hour reading is not near. The 8-hour reading is not near.
|
|
|
|
|
Running total at 00:15Z (8 October): about 2.25 box-hours of run by wall x threads / 96; the lease waits (box 1's line shard s2c has waited since 22:09Z with 0 free adv-class cores) are not counted by wall x threads / 96 (the boxes ran at load 400 to 600 for the first two hours, so the CPU actually consumed is under that). The 8-hour reading is not near. The 8-hour reading is not near.
|
|
|
|
|
|
|
|
|
|
## 6. The bound reached, honestly
|
|
|
|
|
|
|
|
|
|
@ -388,6 +389,7 @@ Not a reason to wait: the 2^35-read census (item 16) and the 16,384-day scan (it
|
|
|
|
|
| 19:49Z | Coordinator's kills (recorded as theirs): on box 2 the resumed census `warps-era-fixed-12-32-2e23` (lease 40 cores, taken 19:41:36Z) ended by SIGTERM after programs 12 to 23 (lost: programs 24 to 32, about 7 min; program 24's rows partial), then the drawn-era census `warps-era-drawn-32-2e23` that the still-alive box 2 drain had started on 40 cores, after programs 1 and 2 (lost: 3 to 32). The order: box 2 closed to adv-* while the class v5 census ran (19:4x to 20:12Z). My 18:2x line "both drains stopped" was wrong for box 2: the lease-aware drain restarted at 19:23:53Z was alive and started the next item on the kill. Box 1 untouched (the 2^35 census ran to its end, 19:35:36Z to 20:05:06Z) |
|
|
|
|
|
| 20:33Z | My own wrong kills on box 1: shard `lines-2e30-s1` killed on purpose (its 64 days duplicated the 256-day census) and, by a `pkill -x` that followed, the shard `s2b` the drain had just started (1 s lost, re-added as `s2c`). Line shards s3b, s4, s2c now cover days 21002 to 21193, fresh |
|
|
|
|
|
| 21:21Z to 21:27Z | Pre-emption rule (a release or v5 waiter pre-empts adv holders at any size after 120 s): the drain re-queues a shard that exits 143, up to 6 times, after the yield test; redeployed on box 1 at 21:23Z (waiting on the running shard s4) and box 2 at 21:27Z (waiting on the running drawn-era shard); each shard's harness logs its rows as it goes |
|
|
|
|
|
| 00:12Z (8 October) | Box 2 queue drained again (the 8 windows shards, 8 minutes): nothing of this lane remains on box 2. Box 1: the line shard s2c still waiting for adv-class cores (since 22:09Z) |
|
|
|
|
|
| 22:23Z | Box 2 queue drained (the drain exited after 6 items; the three adv-cache-3 queue files there are not lease-pool items and were skipped, not run). Box 1: the 8 drawn-era windows shards wait behind the line shard s2c, itself waiting for adv-class cores since 22:09Z; at 00:0xZ the windows shards were copied to box 2 and a drain started there |
|
|
|
|
|
| 20:40Z | My own wrong kill: on box 2 I ended lease pid 2674115 (24 cores, label adv-cache-2 warps-devnet-2e26-v2, held from 20:36:59Z) as an orphan; it was a sibling's script running my old queue item 43 under the sibling-queue rule (claimed by my first drain, superseded by my item 430). About 3 minutes of a sibling's cores lost; reported to the coordinator at 20:42Z |
|
|
|
|
|
| 20:2x to 20:3xZ | Box 2 reopened at 20:12Z. Remaining work re-sharded (lines 2^30 x 4 shards, windows 512 x 8, warps 4 to 5 programs per shard, devnet at 2^25) with every lease at `lease pool 32 --min 16` (holders at 32 or under are never pre-empted under the ranked lease of 19:40Z), the drain holding before each lease while a class-v5 waiter (by owner, never an adv-* lane) is in `lease status`; a first version of that test matched a v5 HOLDER and held box 1 idle from 19:20Z to 20:2xZ (fixed: a waiter line carries "(min N, K free)"); duplicate drains and an orphan 48-core waiter on box 2 killed at 20:2xZ, one drain per box since |
|
|
|
|
|
|
igneum-labs