Skip to content

GitLab

"lib/vscode:/vscode.git/clone" did not exist on "3f4b3d4de961b439d9a2274c2d67fb19e1173f77"
Unverified Commit fb60cdc5 authored by Hongkuan Zhou's avatar Hongkuan Zhou Committed by GitHub
Browse files

fix: profiling data format in disagg_planner.yaml (#7019) 


Signed-off-by: default avatarhongkuanz <hongkuanz@nvidia.com>
parent e77646ae
Show whitespace changes
Inline Side-by-side
Showing with 24 additions and 57 deletions
examples/backends/sglang/deploy/disagg_planner.yaml
View file @ fb60cdc5
...@@ -19,9 +19,10 @@ ...@@ -19,9 +19,10 @@
# prefill_ttft (time-to-first-token in ms), prefill_thpt_per_gpu # prefill_ttft (time-to-first-token in ms), prefill_thpt_per_gpu
# (tokens/s per GPU) — all 1-D arrays of equal length. # (tokens/s per GPU) — all 1-D arrays of equal length.
# decode_raw_data.json — x_kv_usage (KV-cache utilisation fractions), # decode_raw_data.json — x_kv_usage (KV-cache utilisation fractions),
# y_context_length (context lengths), z_itl (inter-token latency in ms, # y_context_length (context lengths), z_itl (inter-token latency in ms),
# shape [len(x_kv_usage), len(y_context_length)]), z_thpt_per_gpu # z_thpt_per_gpu (tokens/s per GPU), max_kv_tokens (scalar).
# (tokens/s per GPU, same shape), max_kv_tokens (scalar). # x_kv_usage, y_context_length, z_itl, and z_thpt_per_gpu must all be
# flat 1-D arrays of the same length (one entry per measurement point).
# ────────────────────────────────────────────────────────────────────────────── # ──────────────────────────────────────────────────────────────────────────────
apiVersion: v1 apiVersion: v1
kind: ConfigMap kind: ConfigMap
...@@ -36,22 +37,10 @@ data: ...@@ -36,22 +37,10 @@ data:
} }
decode_raw_data.json: | decode_raw_data.json: |
{ {
"x_kv_usage": [0.1, 0.3, 0.5, 0.7, 0.9], "x_kv_usage": [0.1, 0.1, 0.1, 0.1, 0.3, 0.3, 0.3, 0.3, 0.5, 0.5, 0.5, 0.5, 0.7, 0.7, 0.7, 0.7, 0.9, 0.9, 0.9, 0.9],
"y_context_length": [128, 512, 1024, 2048], "y_context_length": [128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048],
"z_itl": [ "z_itl": [5, 6, 7, 9, 6, 7, 8, 10, 7, 8, 10, 12, 8, 10, 12, 15, 10, 12, 15, 20],
[5, 6, 7, 9 ], "z_thpt_per_gpu": [4500, 4000, 3500, 2800, 4200, 3700, 3200, 2500, 3800, 3300, 2800, 2200, 3400, 2900, 2400, 1800, 2800, 2400, 1900, 1400],
[6, 7, 8, 10],
[7, 8, 10, 12],
[8, 10, 12, 15],
[10, 12, 15, 20]
],
"z_thpt_per_gpu": [
[4500, 4000, 3500, 2800],
[4200, 3700, 3200, 2500],
[3800, 3300, 2800, 2200],
[3400, 2900, 2400, 1800],
[2800, 2400, 1900, 1400]
],
"max_kv_tokens": 32768 "max_kv_tokens": 32768
} }
--- ---
......
examples/backends/trtllm/deploy/disagg_planner.yaml
View file @ fb60cdc5
...@@ -19,9 +19,10 @@ ...@@ -19,9 +19,10 @@
# prefill_ttft (time-to-first-token in ms), prefill_thpt_per_gpu # prefill_ttft (time-to-first-token in ms), prefill_thpt_per_gpu
# (tokens/s per GPU) — all 1-D arrays of equal length. # (tokens/s per GPU) — all 1-D arrays of equal length.
# decode_raw_data.json — x_kv_usage (KV-cache utilisation fractions), # decode_raw_data.json — x_kv_usage (KV-cache utilisation fractions),
# y_context_length (context lengths), z_itl (inter-token latency in ms, # y_context_length (context lengths), z_itl (inter-token latency in ms),
# shape [len(x_kv_usage), len(y_context_length)]), z_thpt_per_gpu # z_thpt_per_gpu (tokens/s per GPU), max_kv_tokens (scalar).
# (tokens/s per GPU, same shape), max_kv_tokens (scalar). # x_kv_usage, y_context_length, z_itl, and z_thpt_per_gpu must all be
# flat 1-D arrays of the same length (one entry per measurement point).
# ────────────────────────────────────────────────────────────────────────────── # ──────────────────────────────────────────────────────────────────────────────
apiVersion: v1 apiVersion: v1
kind: ConfigMap kind: ConfigMap
...@@ -36,22 +37,10 @@ data: ...@@ -36,22 +37,10 @@ data:
} }
decode_raw_data.json: | decode_raw_data.json: |
{ {
"x_kv_usage": [0.1, 0.3, 0.5, 0.7, 0.9], "x_kv_usage": [0.1, 0.1, 0.1, 0.1, 0.3, 0.3, 0.3, 0.3, 0.5, 0.5, 0.5, 0.5, 0.7, 0.7, 0.7, 0.7, 0.9, 0.9, 0.9, 0.9],
"y_context_length": [128, 512, 1024, 2048], "y_context_length": [128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048],
"z_itl": [ "z_itl": [5, 6, 7, 9, 6, 7, 8, 10, 7, 8, 10, 12, 8, 10, 12, 15, 10, 12, 15, 20],
[5, 6, 7, 9 ], "z_thpt_per_gpu": [4500, 4000, 3500, 2800, 4200, 3700, 3200, 2500, 3800, 3300, 2800, 2200, 3400, 2900, 2400, 1800, 2800, 2400, 1900, 1400],
[6, 7, 8, 10],
[7, 8, 10, 12],
[8, 10, 12, 15],
[10, 12, 15, 20]
],
"z_thpt_per_gpu": [
[4500, 4000, 3500, 2800],
[4200, 3700, 3200, 2500],
[3800, 3300, 2800, 2200],
[3400, 2900, 2400, 1800],
[2800, 2400, 1900, 1400]
],
"max_kv_tokens": 32768 "max_kv_tokens": 32768
} }
--- ---
......
examples/backends/vllm/deploy/disagg_planner.yaml
View file @ fb60cdc5
...@@ -19,9 +19,10 @@ ...@@ -19,9 +19,10 @@
# prefill_ttft (time-to-first-token in ms), prefill_thpt_per_gpu # prefill_ttft (time-to-first-token in ms), prefill_thpt_per_gpu
# (tokens/s per GPU) — all 1-D arrays of equal length. # (tokens/s per GPU) — all 1-D arrays of equal length.
# decode_raw_data.json — x_kv_usage (KV-cache utilisation fractions), # decode_raw_data.json — x_kv_usage (KV-cache utilisation fractions),
# y_context_length (context lengths), z_itl (inter-token latency in ms, # y_context_length (context lengths), z_itl (inter-token latency in ms),
# shape [len(x_kv_usage), len(y_context_length)]), z_thpt_per_gpu # z_thpt_per_gpu (tokens/s per GPU), max_kv_tokens (scalar).
# (tokens/s per GPU, same shape), max_kv_tokens (scalar). # x_kv_usage, y_context_length, z_itl, and z_thpt_per_gpu must all be
# flat 1-D arrays of the same length (one entry per measurement point).
# ────────────────────────────────────────────────────────────────────────────── # ──────────────────────────────────────────────────────────────────────────────
apiVersion: v1 apiVersion: v1
kind: ConfigMap kind: ConfigMap
...@@ -36,22 +37,10 @@ data: ...@@ -36,22 +37,10 @@ data:
} }
decode_raw_data.json: | decode_raw_data.json: |
{ {
"x_kv_usage": [0.1, 0.3, 0.5, 0.7, 0.9], "x_kv_usage": [0.1, 0.1, 0.1, 0.1, 0.3, 0.3, 0.3, 0.3, 0.5, 0.5, 0.5, 0.5, 0.7, 0.7, 0.7, 0.7, 0.9, 0.9, 0.9, 0.9],
"y_context_length": [128, 512, 1024, 2048], "y_context_length": [128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048, 128, 512, 1024, 2048],
"z_itl": [ "z_itl": [5, 6, 7, 9, 6, 7, 8, 10, 7, 8, 10, 12, 8, 10, 12, 15, 10, 12, 15, 20],
[5, 6, 7, 9 ], "z_thpt_per_gpu": [4500, 4000, 3500, 2800, 4200, 3700, 3200, 2500, 3800, 3300, 2800, 2200, 3400, 2900, 2400, 1800, 2800, 2400, 1900, 1400],
[6, 7, 8, 10],
[7, 8, 10, 12],
[8, 10, 12, 15],
[10, 12, 15, 20]
],
"z_thpt_per_gpu": [
[4500, 4000, 3500, 2800],
[4200, 3700, 3200, 2500],
[3800, 3300, 2800, 2200],
[3400, 2900, 2400, 1800],
[2800, 2400, 1900, 1400]
],
"max_kv_tokens": 32768 "max_kv_tokens": 32768
} }
--- ---
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment