mukesh 694e0cc9b9 paper(gqa): add 6-case per-PE memory + comm summary chart ...

Analytical plot generator + PNG covering the 4 sharding cases from
slide 17 plus the 2 d_head-TP variants (Cases 4, 5). Three panels:
per-PE HBM budget breakdown (Wq + Wk + Wv + Wo + 4.24 GB KV
headroom), per-PE KV memory at S_kv=1M, and per-PE communication
per output token (decode). Numbers match slide 17 (max KV context
~7.1M for the 64-way sharded cases).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

2026-06-16 16:12:26 -07:00

..

paper

paper(gqa): add 6-case per-PE memory + comm summary chart

2026-06-16 16:12:26 -07:00

build_overview_slides.py

CCL allreduce: rename to lrab_hierarchical_allreduce + descriptive plots

2026-05-20 20:50:48 -07:00

gemm_sweep.py

eval: fold GEMM/allreduce harnesses into self-contained milestone benches

2026-05-22 15:19:52 -07:00

gen_placement_diagrams.py

commit - release 1

2026-03-18 11:47:48 -07:00

plot_pe_dma_perf.py

PE_DMA perf: SIP-wide scenarios + dual outputs + clearer naming

2026-05-15 09:43:09 -07:00

replot_pe2pe.py

Add tl.recv_no_consume diagnostic API for apples-to-apples pe2pe plot

2026-04-28 18:20:44 -07:00