vh/esh-pfi-infrastructure
1
0
Fork 0
Code Issues 7 Pull Requests Actions Packages Projects Releases Wiki Activity

memory: GPU 0 room can come from trimming gen-small KV (Prime); donor analysis

Browse Source
This commit is contained in:
vh
2026-09-30 16:08:10 -07:00
parent aa17f64bd1
commit 11174ffea1
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
persistent-memory.md
+1 -1
View File
@@ -184,7 +184,7 @@ _As of 2026-09-30 ~0120 PT._
- **Dropout INVESTIGATED 2026-09-30 (Prime via coordinator; investigation only, nothing deployed):** `docs/pfi/parakeet-dropout-investigation-2026-09-30.md`. Real losses against ground truth (SCOTUS official transcript, Gutenberg #38916): v3 loses ~140 / 66 / 50 / 51 clean words per transcript (audiobook / argument / p1 / p2). Cause = v2/v3 0.6B weights collapse deep in long full-attention windows (encoder-side; 1.1B TDT/RNNT/CTC never do). No decoding, context, loudness or resampling fix. **Fix = re-transcribe ≥3 s speech gaps: −80–90 % everywhere** → `stacks/scriberr/patches/proposed/0002` (+ `PARAKEET_MODEL_PATH`), built as `scriberr:local-blackwell-a353078-dropout2`, NOT deployed; peak 5,506 MiB. Prime's calls: ship 0002?; v2 (0 on his files, collapses on read speech) vs keep v3; parakeet-unified-en-0.6b (needs NeMo 3.0.0 + NVIDIA Open Model License). Weights pulled pinned into `/tank/aimodels/huggingface` (~30 GB); throwaway env `/tank/spikes/scriberr-slicer/envs/nemo300`.
- Scriberr moved to **fv-ml1 GPU 3** (coordinator, 2026-09-30); `scriberr-rebuild` memory stage now counts only its own PIDs and needs ≥20 GB free. Its default budget is still the retired GPU 1 5,496 MiB (0002 peaks 5,506 → pass `--budget`).
- Private bench data (copies of Prime's two uploads + transcripts) sits in fv-ml1 `/tank/spikes/scriberr-slicer/private/` (mode 700), kept pending Prime; the public audio and metrics are beside it.
- **Parakeet SEAT A/B IN FLIGHT (Prime 1558: "a/b the one on fv-ml1's general seat against the unified new one"; latency is load-bearing, even 50 ms).** A background agent is comparing the seat (`parakeet`, GPU 0, sherpa-onnx int8 v3, :8300, LiteLLM `ext-stt`/`whisper-1`) with `nvidia/parakeet-unified-en-0.6b`. It runs under NeMo, AND in the seat's own ONNX runtime if an export exists or can be made, with v2 int8 as a cheap English control. Candidates run on GPU 3 only. ⚠ GPU 0 has ~101 MiB free, so an in-place replacement must fit about 1,790 MiB. No deploy.
- **Parakeet SEAT A/B IN FLIGHT (Prime 1558: "a/b the one on fv-ml1's general seat against the unified new one"; latency is load-bearing, even 50 ms).** A background agent is comparing the seat (`parakeet`, GPU 0, sherpa-onnx int8 v3, :8300, LiteLLM `ext-stt`/`whisper-1`) with `nvidia/parakeet-unified-en-0.6b`. It runs under NeMo, AND in the seat's own ONNX runtime if an export exists or can be made, with v2 int8 as a cheap English control. Candidates run on GPU 3 only. ⚠ GPU 0 has ~101 MiB free, but **Prime 1610: room can be freed by trimming KV concurrency on a big GPU 0 seat.** The best donor is `vllm-gen-small` (util 0.48, NVFP4, 670k-token fp8 KV, 2.56× at 262k). Each 0.01 util is about 0.95 GB, or about 35k tokens (~5% of its KV); freeing 2 GB takes it to roughly 2.3×. `vllm-cyberprev` (0.40, 359k KV, 1.37×) would lose ~9% per GB. Do NOT touch `vllm-voices` (0.11). Changing util means a gen-small restart (~2–3 min). The KV figures are from each seat's boot log (09-14); the per-token cost is my estimate, and the restart log will print the real numbers. No deploy.
- **Scriberr gap-retry fix (patch 0002) LIVE 1602 2026-09-30** as `scriberr:local-blackwell-a353078-dropout2` (Prime: "basic fix, no surgery for the new toolkit"; v3 stays). The investigation (`docs/pfi/parakeet-dropout-investigation-2026-09-30.md`) found the v3 drops are real against ground truth (50–140 words per transcript); the retry cuts them 80–90%. Live check: 5,502 MiB, `retried_gaps` reported. Rollback: `.env.bak-20260930-pre-dropout2` / slicer1. Leftovers kept pending Prime: 26 GB of candidate weights, `envs/nemo300`, and the private bench data in `/tank/spikes/scriberr-slicer/`.
- **2026-09-30 1322–1335, Prime: "Go GPU 3 now and extend the jev endpoint to hit 32k tokens".** DONE.
- **Scriberr is on fv-ml1 GPU 3** (`SCRIBERR_GPU_ID=3`; a 20-min file verified at 5,496 MiB). It is an on-demand tenant of the reserve, like Blender: it STEPS ASIDE when a full-size seat claims GPU 3, and it goes to **irv-ml1's A6000**, NOT back to GPU 1.