ywkang baf22f01ab plot: extend batch-scaling curves with saturated-throughput ceiling ...

Beyond a mapping's SIP capacity (16/C) the throughput plateaus: extra
users run in waves at the saturated rate, they are not un-runnable. Draw
a dashed horizontal extension at the ceiling so 1-kv-per-cube (cap 2)
reads as saturating, not stopping, at B=2. Caption updated.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-07-22 17:39:13 -07:00

..

plot_scripts

plot: extend batch-scaling curves with saturated-throughput ceiling

2026-07-22 17:39:13 -07:00

decode_batch_sweep.csv

gqa decode: fix cube_base output-store; measure batch scaling on one SIP

2026-07-22 17:37:10 -07:00

short_context_decode_composite_fused_sweep.csv

gqa short-context: exclude KV deploy from wall; correct HBM peak; batch scaffold

2026-07-22 16:25:04 -07:00

short_context_decode_composite_sweep.csv

gqa short-context: exclude KV deploy from wall; correct HBM peak; batch scaffold

2026-07-22 16:25:04 -07:00

short_context_decode_sweep.4ctx.csv

sweep: 6 short-context CSVs (3 variants x prefill+decode) + 6 wall/util figures

2026-06-26 15:59:21 -07:00

short_context_decode_sweep.csv

gqa short-context: exclude KV deploy from wall; correct HBM peak; batch scaffold

2026-07-22 16:25:04 -07:00

short_context_prefill_composite_fused_sweep.csv

gqa short-context: exclude KV deploy from wall; correct HBM peak; batch scaffold

2026-07-22 16:25:04 -07:00

short_context_prefill_composite_sweep.csv

gqa short-context: exclude KV deploy from wall; correct HBM peak; batch scaffold

2026-07-22 16:25:04 -07:00

short_context_prefill_sweep.4ctx.csv

sweep: 6 short-context CSVs (3 variants x prefill+decode) + 6 wall/util figures

2026-06-26 15:59:21 -07:00

short_context_prefill_sweep.csv

gqa short-context: exclude KV deploy from wall; correct HBM peak; batch scaffold

2026-07-22 16:25:04 -07:00