Datasets:

Hamerlate
/

tts

@@ -65,3 +65,5 @@ qwen_util_bs.nsys-rep filter=lfs diff=lfs merge=lfs -text
 traverse_bs.nsys-rep filter=lfs diff=lfs merge=lfs -text
 sim_traverse_bs/traverse_bs_util_std.nsys-rep filter=lfs diff=lfs merge=lfs -text
 traverse_bs_util_std.nsys-rep filter=lfs diff=lfs merge=lfs -text

 traverse_bs.nsys-rep filter=lfs diff=lfs merge=lfs -text
 sim_traverse_bs/traverse_bs_util_std.nsys-rep filter=lfs diff=lfs merge=lfs -text
 traverse_bs_util_std.nsys-rep filter=lfs diff=lfs merge=lfs -text
+traverse_bs_util_sim_decoding.nsys-rep filter=lfs diff=lfs merge=lfs -text
+traverse_bs_util_sim_prefill.nsys-rep filter=lfs diff=lfs merge=lfs -text

traverse_bs_util_sim_decoding.log ADDED Viewed

	@@ -0,0 +1,390 @@

+WARNING: CPU IP/backtrace sampling not supported, disabling.
+Try the 'nsys status --environment' command to learn more.
+WARNING: CPU context switch tracing not supported, disabling.
+Try the 'nsys status --environment' command to learn more.
+INFO 08-13 19:21:37 [__init__.py:235] Automatically detected platform cuda.
+CUDA_VISIBLE_DEVICES = 3
+--- vLLM V1 基准测试（含 NVTX 标记）---
+模型: Qwen/Qwen2-1.5B
+批量大小: [1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024]
+场景: ['prefill1_decode512']
+------------------------------------------------------------
+加载分词器/模型中...
+INFO 08-13 19:21:46 [config.py:1604] Using max model len 4096
+INFO 08-13 19:21:47 [config.py:2434] Chunked prefill is enabled with max_num_batched_tokens=8192.
+INFO 08-13 19:21:52 [__init__.py:235] Automatically detected platform cuda.
+INFO 08-13 19:21:54 [core.py:572] Waiting for init message from front-end.
+INFO 08-13 19:21:54 [core.py:71] Initializing a V1 LLM engine (v0.10.0) with config: model='Qwen/Qwen2-1.5B', speculative_config=None, tokenizer='Qwen/Qwen2-1.5B', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, override_neuron_config={}, tokenizer_revision=None, trust_remote_code=True, dtype=torch.bfloat16, max_seq_len=4096, download_dir=None, load_format=LoadFormat.AUTO, tensor_parallel_size=1, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto,  device_config=cuda, decoding_config=DecodingConfig(backend='auto', disable_fallback=False, disable_any_whitespace=False, disable_additional_properties=False, reasoning_backend=''), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None), seed=0, served_model_name=Qwen/Qwen2-1.5B, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=True, chunked_prefill_enabled=True, use_async_output_proc=True, pooler_config=None, compilation_config={"level":3,"debug_dump_path":"","cache_dir":"","backend":"","custom_ops":[],"splitting_ops":["vllm.unified_attention","vllm.unified_attention_with_output","vllm.mamba_mixer2"],"use_inductor":true,"compile_sizes":[],"inductor_compile_config":{"enable_auto_functionalized_v2":false},"inductor_passes":{},"use_cudagraph":true,"cudagraph_num_of_warmups":1,"cudagraph_capture_sizes":[512,504,496,488,480,472,464,456,448,440,432,424,416,408,400,392,384,376,368,360,352,344,336,328,320,312,304,296,288,280,272,264,256,248,240,232,224,216,208,200,192,184,176,168,160,152,144,136,128,120,112,104,96,88,80,72,64,56,48,40,32,24,16,8,4,2,1],"cudagraph_copy_inputs":false,"full_cuda_graph":false,"max_capture_size":512,"local_cache_dir":null}
+INFO 08-13 19:21:56 [parallel_state.py:1102] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, TP rank 0, EP rank 0
+WARNING 08-13 19:21:56 [topk_topp_sampler.py:59] FlashInfer is not available. Falling back to the PyTorch-native implementation of top-p & top-k sampling. For the best performance, please install FlashInfer.
+INFO 08-13 19:21:56 [gpu_model_runner.py:1843] Starting to load model Qwen/Qwen2-1.5B...
+INFO 08-13 19:21:56 [gpu_model_runner.py:1875] Loading model from scratch...
+INFO 08-13 19:21:56 [cuda.py:290] Using Flash Attention backend on V1 engine.
+INFO 08-13 19:21:57 [weight_utils.py:296] Using model weights format ['*.safetensors']
+INFO 08-13 19:21:57 [weight_utils.py:349] No model.safetensors.index.json found in remote.
+INFO 08-13 19:21:58 [default_loader.py:262] Loading weights took 0.63 seconds
+INFO 08-13 19:21:58 [gpu_model_runner.py:1892] Model loading took 2.9105 GiB and 1.878581 seconds
+INFO 08-13 19:22:04 [backends.py:530] Using cache directory: /home/cy/.cache/vllm/torch_compile_cache/40b61c71e9/rank_0_0/backbone for vLLM's torch.compile
+INFO 08-13 19:22:04 [backends.py:541] Dynamo bytecode transform time: 5.72 s
+INFO 08-13 19:22:09 [backends.py:161] Directly load the compiled graph(s) for dynamic shape from the cache, took 4.036 s
+INFO 08-13 19:22:10 [monitor.py:34] torch.compile takes 5.72 s in total
+INFO 08-13 19:22:11 [gpu_worker.py:255] Available KV cache memory: 12.81 GiB
+INFO 08-13 19:22:11 [kv_cache_utils.py:833] GPU KV cache size: 479,536 tokens
+INFO 08-13 19:22:11 [kv_cache_utils.py:837] Maximum concurrency for 4,096 tokens per request: 117.07x
+INFO 08-13 19:22:13 [gpu_model_runner.py:2485] Graph capturing finished in 2 secs, took 0.49 GiB
+INFO 08-13 19:22:13 [core.py:193] init engine (profile, create kv cache, warmup model) took 14.62 seconds
+模型加载完成。
+===== 场景：prefill1_decode512 | prefill=1, decode=512 =====
+--- 批量大小 bs=1 ---
+执行时间: 3.3194 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 512
+吞吐(生成tokens/秒): 154.24
+TTFT (V1 metrics): 0.0190 s
+解码吞吐 (V1 metrics): 155.07 tok/s
+--- 批量大小 bs=2 ---
+执行时间: 3.6484 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 1024
+吞吐(生成tokens/秒): 280.67
+TTFT (V1 metrics): 0.0120 s
+解码吞吐 (V1 metrics): 140.73 tok/s
+--- 批量大小 bs=4 ---
+执行时间: 3.6361 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 2048
+吞吐(生成tokens/秒): 563.24
+TTFT (V1 metrics): 0.0140 s
+解码吞吐 (V1 metrics): 141.25 tok/s
+--- 批量大小 bs=8 ---
+执行时间: 3.7267 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 4096
+吞吐(生成tokens/秒): 1099.08
+TTFT (V1 metrics): 0.0149 s
+解码吞吐 (V1 metrics): 137.87 tok/s
+--- 批量大小 bs=16 ---
+执行时间: 3.8260 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 8192
+吞吐(生成tokens/秒): 2141.13
+TTFT (V1 metrics): 0.0136 s
+解码吞吐 (V1 metrics): 134.33 tok/s
+--- 批量大小 bs=32 ---
+执行时间: 3.9972 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 16384
+吞吐(生成tokens/秒): 4098.85
+TTFT (V1 metrics): 0.0164 s
+解码吞吐 (V1 metrics): 128.92 tok/s
+--- 批量大小 bs=64 ---
+执行时间: 4.2731 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 32768
+吞吐(生成tokens/秒): 7668.51
+TTFT (V1 metrics): 0.0198 s
+解码吞吐 (V1 metrics): 120.83 tok/s
+--- 批量大小 bs=128 ---
+执行时间: 4.8421 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 65536
+吞吐(生成tokens/秒): 13534.75
+TTFT (V1 metrics): 0.0316 s
+解码吞吐 (V1 metrics): 107.35 tok/s
+--- 批量大小 bs=256 ---
+执行时间: 7.4408 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 131072
+吞吐(生成tokens/秒): 17615.41
+TTFT (V1 metrics): 0.0433 s
+解码吞吐 (V1 metrics): 69.70 tok/s
+--- 批量大小 bs=512 ---
+执行时间: 12.6794 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 262144
+吞吐(生成tokens/秒): 20674.72
+TTFT (V1 metrics): 0.1809 s
+解码吞吐 (V1 metrics): 42.38 tok/s
+--- 批量大小 bs=1024 ---
+[rank0]:[W813 19:23:32.135663883 ProcessGroupNCCL.cpp:1479] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
+执行时间: 25.2865 s
+实际平均输入 tokens: 1.00（目标 1）
+生成总 tokens: 524288
+吞吐(生成tokens/秒): 20733.89
+TTFT (V1 metrics): 0.1191 s
+解码吞吐 (V1 metrics): 20.77 tok/s
+完成。提示：在 Nsight Systems 中可通过 NVTX 区间快速定位各场景/批量的调用。
+GPU 3: General Metrics for NVIDIA AD10x (any frequency)
+Generating '/tmp/nsys-report-bd68.qdstrm'
+[3/8] Executing 'nvtx_sum' stats report
+ Time (%)  Total Time (ns)  Instances      Avg (ns)          Med (ns)         Min (ns)        Max (ns)     StdDev (ns)   Style                   Range
+ --------  ---------------  ---------  ----------------  ----------------  --------------  --------------  -----------  -------  --------------------------------------
+     31.5   35,202,819,763          1  35,202,819,763.0  35,202,819,763.0  35,202,819,763  35,202,819,763          0.0  PushPop  :LLM_init
+     22.6   25,285,906,551          1  25,285,906,551.0  25,285,906,551.0  25,285,906,551  25,285,906,551          0.0  PushPop  :generate [prefill1_decode512] bs=1024
+     11.3   12,679,311,252          1  12,679,311,252.0  12,679,311,252.0  12,679,311,252  12,679,311,252          0.0  PushPop  :generate [prefill1_decode512] bs=512
+      6.7    7,440,608,085          1   7,440,608,085.0   7,440,608,085.0   7,440,608,085   7,440,608,085          0.0  PushPop  :generate [prefill1_decode512] bs=256
+      4.3    4,841,914,697          1   4,841,914,697.0   4,841,914,697.0   4,841,914,697   4,841,914,697          0.0  PushPop  :generate [prefill1_decode512] bs=128
+      3.8    4,272,889,441          1   4,272,889,441.0   4,272,889,441.0   4,272,889,441   4,272,889,441          0.0  PushPop  :generate [prefill1_decode512] bs=64
+      3.6    3,997,075,015          1   3,997,075,015.0   3,997,075,015.0   3,997,075,015   3,997,075,015          0.0  PushPop  :generate [prefill1_decode512] bs=32
+      3.4    3,825,710,172          1   3,825,710,172.0   3,825,710,172.0   3,825,710,172   3,825,710,172          0.0  PushPop  :generate [prefill1_decode512] bs=16
+      3.3    3,726,603,655          1   3,726,603,655.0   3,726,603,655.0   3,726,603,655   3,726,603,655          0.0  PushPop  :generate [prefill1_decode512] bs=8
+      3.3    3,648,294,896          1   3,648,294,896.0   3,648,294,896.0   3,648,294,896   3,648,294,896          0.0  PushPop  :generate [prefill1_decode512] bs=2
+      3.2    3,635,960,724          1   3,635,960,724.0   3,635,960,724.0   3,635,960,724   3,635,960,724          0.0  PushPop  :generate [prefill1_decode512] bs=4
+      3.0    3,319,210,677          1   3,319,210,677.0   3,319,210,677.0   3,319,210,677   3,319,210,677          0.0  PushPop  :generate [prefill1_decode512] bs=1
+      0.0           90,630          2          45,315.0          45,315.0          41,468          49,162      5,440.5  PushPop  CCCL:cub::DeviceSegmentedRadixSort
+[4/8] Executing 'osrt_sum' stats report
+ Time (%)   Total Time (ns)   Num Calls     Avg (ns)          Med (ns)      Min (ns)      Max (ns)       StdDev (ns)              Name
+ --------  -----------------  ---------  ---------------  ----------------  ---------  --------------  ----------------  ----------------------
+     29.7  1,284,023,267,442     49,709     25,830,800.6          28,820.0      1,000  96,534,138,823     969,651,617.8  pthread_cond_timedwait
+     24.2  1,045,773,118,476     73,135     14,299,215.4      10,062,843.0      1,010  81,708,518,315     463,992,514.8  epoll_wait
+     23.9  1,031,963,983,489        549  1,879,715,816.9          15,827.0      1,644  96,535,575,076  13,212,126,179.1  pthread_cond_wait
+      8.3    357,289,068,553         57  6,268,229,272.9  10,000,073,611.0     10,419  10,000,146,360   4,730,020,985.8  sem_timedwait
+      8.2    355,562,427,280     39,343      9,037,501.6           1,512.0      1,000  12,219,368,064     127,255,975.8  poll
+      3.0    131,715,516,186     11,467     11,486,484.4       7,170,376.0     28,456     585,263,318      14,022,810.8  sem_wait
+      2.6    112,179,081,690     41,286      2,717,121.6           2,213.0      1,000  94,667,308,194     468,888,772.5  read
+      0.0        793,423,700        330      2,404,314.2       1,354,549.5      1,900      18,406,079       2,639,127.9  pthread_rwlock_wrlock
+      0.0        494,035,258    199,029          2,482.2           1,380.0      1,000      72,123,224         161,811.1  munmap
+      0.0        298,099,876      8,608         34,630.6          10,089.5      1,002      29,694,619         390,579.6  ioctl
+      0.0        220,903,685        369        598,655.0           2,510.0      1,159      22,536,588       3,264,985.3  fopen
+      0.0        121,576,605         24      5,065,691.9       5,064,718.5  5,053,737       5,087,479           7,589.0  nanosleep
+      0.0        110,429,690     30,645          3,603.5           2,536.0      1,000      19,587,041         111,890.3  open64
+      0.0         88,325,670         79      1,118,046.5           3,103.0      1,011      81,521,546       9,166,201.0  waitpid
+      0.0         76,471,088     18,154          4,212.4           3,660.0      1,000       1,659,563          15,417.3  mmap64
+      0.0         74,586,704         96        776,944.8           3,874.0      1,020      19,635,272       3,727,062.1  open
+      0.0         72,521,171      8,897          8,151.2           4,707.0      1,022       2,826,985          34,660.7  recv
+      0.0         71,841,066      8,895          8,076.6           5,429.0      1,571          84,986           7,239.5  send
+      0.0         69,801,955     41,067          1,699.7           1,627.0      1,000          32,060             794.7  pthread_cond_signal
+      0.0         67,085,379         39      1,720,137.9         470,979.0      3,544      10,373,042       3,340,509.2  pthread_join
+      0.0         56,617,564         10      5,661,756.4          18,705.5      8,315      56,388,994      17,823,747.2  connect
+      0.0         51,207,160     14,809          3,457.8           2,380.0      1,013         139,725           5,733.7  write
+      0.0         40,211,592      4,773          8,424.8           6,319.0      1,000         661,710          13,119.1  pthread_mutex_lock
+      0.0         16,225,859     10,123          1,602.9           1,387.0      1,000          17,416             730.6  epoll_ctl
+      0.0          9,852,733        147         67,025.4          68,737.0     55,805          95,256           5,155.2  sleep
+      0.0          7,858,705         22        357,213.9         474,706.5      8,796         678,261         278,233.7  pthread_rwlock_rdlock
+      0.0          7,721,440        131         58,942.3          56,260.0     21,296         195,560          26,034.2  pthread_create
+      0.0          7,224,126        929          7,776.2           3,096.0      1,000          86,864          11,142.9  fgets
+      0.0          1,723,609        344          5,010.5           4,755.0      1,827          40,649           2,516.1  fopen64
+      0.0          1,708,972         62         27,564.1           2,983.5      1,002         230,421          59,773.0  futex
+      0.0          1,347,355      1,069          1,260.4           1,023.0      1,000          12,904             880.3  fclose
+      0.0          1,149,466        196          5,864.6           3,579.5      1,105         168,420          13,582.2  mmap
+      0.0            878,967          1        878,967.0         878,967.0    878,967         878,967               0.0  fork
+      0.0            364,215         65          5,603.3           5,028.0      1,909          15,104           3,123.9  pipe2
+      0.0            247,833         41          6,044.7           4,941.0      1,709          17,457           4,172.6  socket
+      0.0            188,362         19          9,913.8           3,097.0      1,045          62,742          16,639.4  bind
+      0.0            128,433         34          3,777.4           3,261.0      1,187          14,840           2,461.4  pthread_cond_broadcast
+      0.0             76,747          7         10,963.9           9,959.0      3,576          31,262           9,493.1  fread
+      0.0             65,399         41          1,595.1           1,200.0      1,012           5,988           1,063.6  fcntl
+      0.0             49,079          5          9,815.8           9,542.0      4,750          17,158           4,761.0  accept4
+      0.0             42,725         25          1,709.0           1,806.0      1,011           2,296             397.2  sigaction
+      0.0             40,441         20          2,022.1           2,166.5      1,063           3,618             818.8  dup2
+      0.0             39,878         15          2,658.5           2,065.0      1,267           7,040           1,459.6  stat
+      0.0             31,245         12          2,603.8           1,918.0      1,006           5,220           1,771.4  fflush
+      0.0             27,179          5          5,435.8           5,277.0      1,662           9,374           3,035.3  fwrite
+      0.0             21,540          4          5,385.0           5,545.5      4,572           5,877             575.6  lstat
+      0.0             17,255          4          4,313.8           4,516.5      2,856           5,366           1,051.1  flock
+      0.0             16,827          9          1,869.7           1,599.0      1,008           3,313             844.3  pread
+      0.0             15,569         10          1,556.9           1,444.0      1,184           2,260             325.7  listen
+      0.0             13,074          3          4,358.0           4,294.0      4,285           4,495             118.7  fputs_unlocked
+      0.0             12,439          5          2,487.8           2,713.0      1,831           3,023             566.6  mprotect
+      0.0              7,489          4          1,872.3           1,856.5      1,636           2,140             206.8  flockfile
+      0.0              6,919          1          6,919.0           6,919.0      6,919           6,919               0.0  kill
+      0.0              5,460          2          2,730.0           2,730.0      2,008           3,452           1,021.1  openat64
+      0.0              5,297          3          1,765.7           1,842.0      1,157           2,298             574.3  fstat
+      0.0              3,627          1          3,627.0           3,627.0      3,627           3,627               0.0  fputs
+[5/8] Executing 'cuda_api_sum' stats report
+ Time (%)  Total Time (ns)  Num Calls   Avg (ns)     Med (ns)    Min (ns)   Max (ns)    StdDev (ns)                     Name
+ --------  ---------------  ---------  -----------  -----------  --------  -----------  -----------  ------------------------------------------
+     65.2   20,318,348,211     12,196  1,665,984.6      3,806.0     1,713  143,905,820  4,877,424.6  cudaStreamSynchronize
+     19.6    6,114,479,110    979,090      6,245.1      4,893.0       826   61,317,707    103,504.7  cudaLaunchKernel
+      5.9    1,844,232,020    151,960     12,136.3      9,991.0     7,437    6,397,280     55,443.5  cudaGraphLaunch_v10000
+      4.4    1,366,270,456     61,593     22,182.2      8,605.0     2,898   97,438,741    427,651.1  cudaMemcpyAsync
+      2.1      657,913,070    123,014      5,348.3      4,791.0       646   11,218,430     74,172.4  cuLaunchKernel
+      0.7      225,191,175      1,943    115,898.7     75,223.0    40,921    1,507,774    191,247.4  cudaGraphInstantiateWithFlags_v11040
+      0.6      190,028,321      2,135     89,006.2     32,930.0     5,778  121,430,749  2,627,383.4  cudaDeviceSynchronize
+      0.4      131,733,072     24,728      5,327.3      5,346.0       183    7,263,692     48,804.8  cudaMemsetAsync
+      0.4      117,166,497    154,261        759.5        737.0       297        9,955        164.6  cudaStreamIsCapturing_v10000
+      0.2       54,817,946        222    246,927.7    125,544.5    64,964    2,389,846    359,442.8  cudaFree
+      0.1       41,574,143        348    119,465.9    111,957.5     6,496    1,314,648     70,124.0  cudaMalloc
+      0.1       25,470,407         10  2,547,040.7  2,568,202.0    60,182    4,674,895  1,473,736.1  cuLibraryLoadData
+      0.0       14,126,639     13,502      1,046.3        512.0       267    4,070,645     36,594.4  cuKernelGetFunction
+      0.0       11,511,739        169     68,116.8     73,800.0    26,538      398,968     40,288.3  cuModuleLoadData
+      0.0        9,477,345     18,895        501.6        477.0       305        7,151        120.4  cudaStreamGetCaptureInfo_v2_v11030
+      0.0        8,547,159      1,943      4,398.9      4,349.0     3,274       12,306        644.1  cudaStreamBeginCapture_v10000
+      0.0        7,583,507      1,943      3,903.0      3,886.0     2,371       10,115        530.0  cudaGraphDestroy_v10000
+      0.0        2,953,354        128     23,073.1      2,127.0     1,339      976,651    118,496.6  cudaStreamCreateWithPriority
+      0.0        2,583,759      1,943      1,329.8      1,322.0       973        2,362        129.3  cudaStreamEndCapture_v10000
+      0.0        1,910,887         26     73,495.7     12,773.5     3,625    1,207,162    232,915.8  cudaHostAlloc
+      0.0        1,625,828      1,943        836.8        771.0       625        3,016        254.6  cudaGraphGetNodes_v10000
+      0.0          943,862        310      3,044.7      2,639.0       879       11,991      1,944.4  cudaEventQuery
+      0.0          731,374        311      2,351.7      2,439.0       991        7,657      1,133.3  cudaEventRecord
+      0.0          219,541          8     27,442.6     26,305.5     8,804       64,233     18,995.2  cudaMemGetInfo
+      0.0          140,500        810        173.5        143.0        85        1,704        110.6  cuGetProcAddress_v2
+      0.0           21,914         21      1,043.5        438.0       339        4,729      1,202.9  cudaEventCreateWithFlags
+      0.0           16,258         16      1,016.1        849.5       502        2,663        551.3  cuLibraryGetKernel
+      0.0            8,991         14        642.2        586.0       346        1,420        261.0  cudaThreadExchangeStreamCaptureMode_v10010
+      0.0            4,849          3      1,616.3      1,664.0     1,386        1,799        210.6  cuInit
+      0.0            3,460          4        865.0        749.0       110        1,852        882.7  cuModuleGetLoadingMode
+      0.0            3,416          1      3,416.0      3,416.0     3,416        3,416          0.0  cudaStreamWaitEvent
+      0.0            1,901          1      1,901.0      1,901.0     1,901        1,901          0.0  cudaEventDestroy
+      0.0            1,166          2        583.0        583.0       248          918        473.8  cudaGetDriverEntryPoint_v11030
+[6/8] Executing 'cuda_gpu_kern_sum' stats report
+ Time (%)  Total Time (ns)  Instances   Avg (ns)     Med (ns)    Min (ns)   Max (ns)   StdDev (ns)                                                  Name
+ --------  ---------------  ---------  -----------  -----------  ---------  ---------  -----------  ----------------------------------------------------------------------------------------------------
+     33.5    9,507,682,829     84,588    112,399.9     58,880.0      5,728    569,477    137,306.6  void flash::flash_fwd_splitkv_kernel<Flash_fwd_kernel_traits<(int)128, (int)64, (int)128, (int)4, (…
+     27.2    7,733,720,607     29,164    265,180.4    333,123.0     33,344    763,622    112,810.5  ampere_bf16_s1688gemm_bf16_64x128_sliced1x2_ldg8_f2f_tn
+      7.4    2,089,759,636      1,164  1,795,326.1  1,390,859.0     40,065  4,518,698  1,024,759.1  ampere_bf16_s1688gemm_bf16_128x128_ldg8_f2f_stages_32x1_tn
+      3.3      942,739,585      5,754    163,840.7     13,376.0      1,951  1,008,234    293,922.8  void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIterator…
+      3.3      942,069,183     76,664     12,288.3      8,032.0      6,240     73,248      8,482.6  void flash::flash_fwd_splitkv_kernel<Flash_fwd_kernel_traits<(int)128, (int)64, (int)128, (int)4, (…
+      3.3      926,915,977      5,958    155,575.0     73,120.5      7,649    549,540    194,687.6  void cutlass::Kernel2<cutlass_80_tensorop_bf16_s16816gemm_relu_bf16_64x64_32x6_tn_align8>(T1::Param…
+      2.8      781,701,830      1,991    392,617.7    496,547.0     10,528    506,724    194,713.8  void cutlass::Kernel2<cutlass_80_wmma_tensorop_bf16_s161616gemm_bf16_16x16_128x2_tn_align8>(T1::Par…
+      2.5      718,309,736      5,756    124,793.2      9,920.0      5,151    716,420    213,447.3  void at::native::reduce_kernel<(int)512, (int)1, at::native::ReduceOp<float, at::native::ArgMaxOps<…
+      2.4      679,170,209    292,768      2,319.8      1,889.0      1,631      6,304        962.1  void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::n…
+      2.1      605,586,081     14,252     42,491.3     42,529.0     26,240     62,817      1,642.0  ampere_bf16_s1688gemm_bf16_128x64_sliced1x2_ldg8_relu_f2f_tn
+      1.8      516,743,427     13,776     37,510.4     37,472.0     36,608     42,560        332.8  ampere_bf16_s1688gemm_bf16_64x64_sliced1x4_ldg8_f2f_tn
+      1.3      366,625,152     16,268     22,536.6     23,936.0      1,055    462,659     19,521.4  triton_poi_fused_mul_silu_1
+      1.2      345,171,402        112  3,081,887.5  3,078,316.5  3,036,028  3,128,477     29,283.9  void flash::flash_fwd_splitkv_kernel<Flash_fwd_kernel_traits<(int)128, (int)64, (int)128, (int)4, (…
+      0.9      260,532,523        513    507,860.7    507,843.0    506,403    509,475        427.2  void cutlass::Kernel2<cutlass_80_wmma_tensorop_bf16_s161616gemm_bf16_16x16_128x1_tn_align8>(T1::Par…
+      0.9      255,622,286        604    423,215.7    487,970.0      7,008    488,866    160,519.3  std::enable_if<!T7, void>::type internal::gemvx::kernel<int, int, __nv_bfloat16, __nv_bfloat16, __n…
+      0.9      242,493,430    161,056      1,505.6      1,280.0      1,023      3,488        458.1  void vllm::reshape_and_cache_flash_kernel<__nv_bfloat16, __nv_bfloat16, (vllm::Fp8KVCacheDataType)0…
+      0.7      206,425,050    146,384      1,410.2      1,344.0      1,183      2,208        221.6  void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.7      203,499,133        184  1,105,973.5    579,541.0    369,795  2,808,909    981,858.0  ampere_bf16_s16816gemm_bf16_128x64_ldg8_f2f_tn
+      0.7      187,786,954      1,120    167,666.9    158,929.0     40,416  1,415,463    206,084.0  ampere_bf16_s1688gemm_bf16_128x64_sliced1x2_ldg8_f2f_tn
+      0.6      180,432,004     16,268     11,091.2     11,936.0      1,505    111,617      4,905.8  triton_red_fused__to_copy_add_mean_mul_pow_rsqrt_2
+      0.5      135,678,243     43,792      3,098.2      3,104.0      2,943      3,616         73.3  void flash::flash_fwd_splitkv_combine_kernel<Flash_fwd_kernel_traits<(int)128, (int)64, (int)128, (…
+      0.3       98,055,325     32,872      2,982.9      2,945.0      2,847      3,233         98.3  void flash::flash_fwd_splitkv_combine_kernel<Flash_fwd_kernel_traits<(int)128, (int)64, (int)128, (…
+      0.3       83,656,872     16,268      5,142.4      5,472.0      1,536     79,136      3,233.8  triton_red_fused__to_copy_add_mean_mul_pow_rsqrt_0
+      0.2       52,052,585     15,687      3,318.2      3,457.0      1,344     22,048        905.0  triton_poi_fused_cat_3
+      0.1       39,252,777          8  4,906,597.1  4,863,769.0  4,802,745  5,085,370    117,043.6  void at_cuda_detail::cub::DeviceSegmentedRadixSortKernel<at_cuda_detail::cub::DeviceRadixSortPolicy…
+      0.1       35,101,287     15,687      2,237.6      2,336.0        863     16,672        687.2  triton_poi_fused_view_5
+      0.1       26,987,031     17,256      1,563.9      1,408.0      1,023      2,784        465.1  void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIterator…
+      0.1       22,634,043        784     28,870.0     12,543.5     11,616     62,720     20,595.1  ampere_bf16_s16816gemm_bf16_64x64_ldg8_f2f_stages_64x5_tn
+      0.1       22,584,985     15,687      1,439.7      1,440.0      1,215      6,720        194.4  triton_poi_fused_cat_4
+      0.1       20,606,081      5,888      3,499.7      3,136.0      2,687      7,488        970.6  void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void …
+      0.1       20,451,721          4  5,112,930.3  5,112,153.5  4,937,754  5,289,660    201,308.2  void at_cuda_detail::cub::DeviceSegmentedRadixSortKernel<at_cuda_detail::cub::DeviceRadixSortPolicy…
+      0.1       15,066,944      5,752      2,619.4      2,368.0      1,952      3,969        554.4  void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void …
+      0.1       14,293,381         28    510,477.9    512,002.5    468,706    513,474      8,208.5  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<signed char>, std::a…
+      0.0        9,733,874          4  2,433,468.5  2,435,244.5  2,367,692  2,495,693     60,628.3  void at::native::<unnamed>::cunn_SoftMaxForward<(int)4, float, float, float, at::native::<unnamed>:…
+      0.0        9,136,049         28    326,287.5    326,210.0    324,514    329,538        982.3  ampere_bf16_s1688gemm_bf16_128x128_ldg8_relu_f2f_stages_32x1_tn
+      0.0        8,425,678        224     37,614.6     37,568.5     36,608     38,880        409.5  void cutlass::Kernel2<cutlass_80_tensorop_bf16_s16816gemm_relu_bf16_64x128_32x6_tn_align8>(T1::Para…
+      0.0        7,777,256          2  3,888,628.0  3,888,628.0  3,705,715  4,071,541    258,678.0  void at::native::_scatter_gather_elementwise_kernel<(int)128, (int)8, void at::native::_cuda_scatte…
+      0.0        7,758,429      8,970        864.9        864.0        767      1,280         77.7  void at::native::vectorized_elementwise_kernel<(int)2, at::native::FillFunctor<long>, std::array<ch…
+      0.0        7,519,688      5,754      1,306.9      1,152.0      1,023      2,048        279.6  void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIterator…
+      0.0        7,231,396        336     21,522.0     21,440.0     21,056     22,592        362.7  ampere_bf16_s16816gemm_bf16_128x64_ldg8_relu_f2f_stages_64x3_tn
+      0.0        6,130,365        476     12,878.9     12,800.5     11,744     14,432        645.4  ampere_bf16_s16816gemm_bf16_64x64_ldg8_relu_f2f_stages_64x5_tn
+      0.0        5,896,061          4  1,474,015.3  1,473,463.0  1,473,191  1,475,944      1,292.3  void at::native::vectorized_elementwise_kernel<(int)4, at::native::<unnamed>::masked_fill_kernel(at…
+      0.0        5,380,367      5,752        935.4        896.0        863      1,344         76.3  void at::native::unrolled_elementwise_kernel<at::native::CUDAFunctorOnSelf_add<int>, std::array<cha…
+      0.0        4,603,692      5,292        869.9        864.0        767      1,185         33.2  void at::native::unrolled_elementwise_kernel<at::native::FillFunctor<int>, std::array<char *, (unsi…
+      0.0        3,996,949          2  1,998,474.5  1,998,474.5  1,996,490  2,000,459      2,806.5  void at::native::vectorized_elementwise_kernel<(int)4, at::native::BinaryFunctor<float, float, floa…
+      0.0        3,842,156      4,143        927.4        896.0        800      1,856         93.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<int>, std::array<cha…
+      0.0        3,593,747         56     64,174.1     64,144.5     63,105     65,728        478.5  void cutlass::Kernel2<cutlass_80_wmma_tensorop_bf16_s161616gemm_bf16_32x32_64x1_tn_align8>(T1::Para…
+      0.0        3,433,971          4    858,492.8    858,901.0    855,877    860,292      2,168.9  void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.0        3,191,215          2  1,595,607.5  1,595,607.5  1,560,359  1,630,856     49,848.9  void at::native::tensor_kernel_scan_innermost_dim<float, std::plus<float>>(T1 *, const T1 *, unsign…
+      0.0        2,871,002      1,512      1,898.8      1,760.0      1,312      2,912        445.3  void cublasLt::splitKreduce_kernel<(int)32, (int)16, int, __nv_bfloat16, __nv_bfloat16, float, (boo…
+      0.0        2,643,071        581      4,549.2      4,671.0      1,984     36,256      1,427.2  triton_red_fused__to_copy_add_embedding_mean_mul_pow_rsqrt_0
+      0.0        2,581,742          2  1,290,871.0  1,290,871.0  1,290,663  1,291,079        294.2  at::native::<unnamed>::fill_reverse_indices_kernel(long *, int, at::cuda::detail::IntDivider<unsign…
+      0.0        2,581,389          2  1,290,694.5  1,290,694.5  1,290,406  1,290,983        408.0  void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.0        2,421,998        112     21,625.0     21,552.0      9,408     34,465     12,020.4  void cutlass::Kernel2<cutlass_80_wmma_tensorop_bf16_s161616gemm_bf16_32x32_128x2_tn_align8>(T1::Par…
+      0.0        1,835,794        581      3,159.7      3,200.0      1,632     39,200      1,543.9  triton_poi_fused_cat_1
+      0.0        1,365,128          2    682,564.0    682,564.0    677,764    687,364      6,788.2  void at::native::<unnamed>::distribution_elementwise_grid_stride_kernel<float, (int)4, void at::nat…
+      0.0        1,304,994        581      2,246.1      2,368.0        863     14,272        629.4  triton_poi_fused_view_3
+      0.0        1,202,982         56     21,481.8     21,456.0     21,152     21,888        275.5  ampere_bf16_s16816gemm_bf16_128x64_ldg8_f2f_stages_32x6_tn
+      0.0        1,027,854      1,153        891.5        896.0        800      1,216         36.7  void at::native::vectorized_elementwise_kernel<(int)2, at::native::FillFunctor<int>, std::array<cha…
+      0.0          956,098         28     34,146.4     34,736.5     17,920     35,200      3,188.0  std::enable_if<!T7, void>::type internal::gemvx::kernel<int, int, __nv_bfloat16, float, float, floa…
+      0.0          847,695        581      1,459.0      1,440.0      1,216      9,408        336.4  triton_poi_fused_cat_2
+      0.0          611,794        673        909.1        896.0        864      1,025         28.1  void at::native::unrolled_elementwise_kernel<at::native::FillFunctor<long>, std::array<char *, (uns…
+      0.0          417,574        308      1,355.8      1,344.0      1,311      1,504         21.9  void vllm::merge_attn_states_kernel<__nv_bfloat16, (unsigned int)128>(T1 *, float *, const T1 *, co…
+      0.0          295,335        168      1,757.9      1,760.0      1,535      2,080        119.1  void cublasLt::splitKreduce_kernel<(int)32, (int)16, int, __nv_bfloat16, __nv_bfloat16, float, (boo…
+      0.0          155,841          1    155,841.0    155,841.0    155,841    155,841          0.0  void at::native::<unnamed>::CatArrayBatchedCopy_aligned16_contig<at::native::<unnamed>::OpaqueType<…
+      0.0           78,880          1     78,880.0     78,880.0     78,880     78,880          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::bfloat16_copy_kernel_cuda(at::Te…
+      0.0           63,740         58      1,099.0        896.0        864     11,360      1,372.6  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<c10::BFloat16>, std:…
+      0.0           43,936          1     43,936.0     43,936.0     43,936     43,936          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::sin_kernel_cuda(at::TensorIterat…
+      0.0           36,570         28      1,306.1      1,312.0      1,280      1,376         19.5  void cublasLt::splitKreduce_kernel<(int)32, (int)16, int, float, __nv_bfloat16, float, (bool)0, __n…
+      0.0           26,816          1     26,816.0     26,816.0     26,816     26,816          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::cos_kernel_cuda(at::TensorIterat…
+      0.0           19,520          1     19,520.0     19,520.0     19,520     19,520          0.0  void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.0           11,936         11      1,085.1        864.0        864      1,568        286.4  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<float>, std::array<c…
+      0.0           10,752          2      5,376.0      5,376.0      5,120      5,632        362.0  void at::native::_scatter_gather_elementwise_kernel<(int)128, (int)8, void at::native::_cuda_scatte…
+      0.0            9,152          2      4,576.0      4,576.0      4,480      4,672        135.8  void at::native::<unnamed>::distribution_elementwise_grid_stride_kernel<float, (int)4, void at::nat…
+      0.0            3,616          2      1,808.0      1,808.0      1,600      2,016        294.2  void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.0            3,424          2      1,712.0      1,712.0      1,664      1,760         67.9  void at::native::vectorized_elementwise_kernel<(int)2, at::native::CUDAFunctorOnOther_add<long>, st…
+      0.0            3,136          2      1,568.0      1,568.0      1,504      1,632         90.5  void at::native::vectorized_elementwise_kernel<(int)2, at::native::<unnamed>::where_kernel_impl(at:…
+      0.0            3,104          2      1,552.0      1,552.0      1,344      1,760        294.2  void at::native::vectorized_elementwise_kernel<(int)4, void at::native::compare_scalar_kernel<float…
+      0.0            2,975          2      1,487.5      1,487.5        992      1,983        700.7  void <unnamed>::elementwise_kernel_with_index<int, at::native::arange_cuda_out(const c10::Scalar &,…
+      0.0            2,912          2      1,456.0      1,456.0      1,344      1,568        158.4  void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnOther_add<float>, s…
+      0.0            2,336          1      2,336.0      2,336.0      2,336      2,336          0.0  void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::…
+      0.0            1,184          1      1,184.0      1,184.0      1,184      1,184          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::reciprocal_kernel_cuda(at::Tenso…
+      0.0            1,024          1      1,024.0      1,024.0      1,024      1,024          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<float, float, floa…
+      0.0            1,024          1      1,024.0      1,024.0      1,024      1,024          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<float, float, floa…
+      0.0              896          1        896.0        896.0        896        896          0.0  void at::native::vectorized_elementwise_kernel<(int)2, at::native::FillFunctor<double>, std::array<…
+[7/8] Executing 'cuda_gpu_mem_time_sum' stats report
+ Time (%)  Total Time (ns)  Count   Avg (ns)  Med (ns)  Min (ns)   Max (ns)   StdDev (ns)            Operation
+ --------  ---------------  ------  --------  --------  --------  ----------  -----------  ------------------------------
+     93.2      540,571,743  41,277  13,096.2     352.0       287  97,068,545    513,408.1  [CUDA memcpy Host-to-Device]
+      3.2       18,710,334  14,564   1,284.7     896.0       864   1,362,855     22,521.7  [CUDA memcpy Device-to-Device]
+      2.5       14,536,294  21,760     668.0     768.0       287       7,744        311.5  [CUDA memset]
+      1.1        6,503,130   5,752   1,130.6   1,120.0       863       1,760         95.6  [CUDA memcpy Device-to-Host]
+[8/8] Executing 'cuda_gpu_mem_size_sum' stats report
+ Total (MB)  Count   Avg (MB)  Med (MB)  Min (MB)  Max (MB)  StdDev (MB)            Operation
+ ----------  ------  --------  --------  --------  --------  -----------  ------------------------------
+  4,190.741  41,277     0.102     0.000     0.000   466.747        2.619  [CUDA memcpy Host-to-Device]
+  2,534.048  14,564     0.174     0.003     0.003   622.330       10.312  [CUDA memcpy Device-to-Device]
+     14.589  21,760     0.001     0.001     0.000     0.006        0.000  [CUDA memset]
+      4.192   5,752     0.001     0.000     0.000     0.004        0.001  [CUDA memcpy Device-to-Host]
+Generated:
+    /data/cy/kv_cache_vs_util/sim_traverse_bs/traverse_bs_util_sim_decoding.nsys-rep
+    /data/cy/kv_cache_vs_util/sim_traverse_bs/traverse_bs_util_sim_decoding.sqlite

traverse_bs_util_sim_decoding.nsys-rep ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f9cb1e2cf07fcb0e9c21d44c66cdf7762bc995625a1a55e835b8cffc5a6104b8
+size 109416019

traverse_bs_util_sim_decoding.py ADDED Viewed

	@@ -0,0 +1,274 @@

+import os
+import time
+import statistics
+from typing import List, Tuple, Dict
+import torch
+import torch.cuda.nvtx as nvtx
+from vllm import LLM, SamplingParams
+from transformers import AutoTokenizer
+# ========= 强制使用 vLLM V1 =========
+os.environ.setdefault("VLLM_USE_V1", "1")
+os.environ.setdefault("VLLM_WORKER_MULTIPROC_METHOD", "spawn")
+# 可选：打开 V1 metrics 统计
+os.environ.setdefault("VLLM_LOGGING_LEVEL", "INFO")
+# ========= 试图导入 V1 metrics 类型（兼容不同版本）=========
+try:
+    from vllm.v1.metrics.reader import Counter, Gauge, Histogram, Vector  # type: ignore
+except Exception:
+    Counter = Gauge = Histogram = Vector = type("X", (), {})  # dummy
+# ========= 配置 =========
+MODEL_NAME = "Qwen/Qwen2-1.5B"
+DTYPE = "bfloat16"
+TP = 1
+GPU_MEM_UTIL = 0.90
+TRUST_REMOTE_CODE = True
+# 场景：prefill=输入tokens，decode=输出tokens
+SCENARIOS = [
+    # {"name": "prefill640_decode1", "prompt_tokens": 640, "max_new_tokens": 1},
+    {"name": "prefill1_decode512", "prompt_tokens": 1,   "max_new_tokens": 512},
+    # {"name": "prefill640_decode512", "prompt_tokens": 640, "max_new_tokens": 512},
+]
+BATCH_SIZES = [1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024]
+SEED = 1234
+TEMPERATURE = 0.0
+TOP_P = 1.0
+WARMUP_PER_BS = 1  # 每个批次做一次预热
+# ========= 构造“精确 token 数量”的 prompt =========
+def build_exact_token_prompt(tokenizer, target_len: int) -> str:
+    if target_len <= 1:
+        # 最小化 prompt：用一个简单 token（避免空串导致0 token）
+        ids = tokenizer("A", add_special_tokens=False)["input_ids"]
+        if len(ids) >= 1:
+            return tokenizer.decode(ids[:1], skip_special_tokens=True, clean_up_tokenization_spaces=False)
+    base_text = (
+        "You are a helpful assistant. "
+        "Please analyze the following input and respond succinctly. "
+    )
+    chunk = " ".join(["data"] * 100) + ". "
+    text = base_text + chunk * 200  # 足够长的文本
+    lo, hi = 0, len(text)
+    target_ids = None
+    while lo <= hi:
+        mid = (lo + hi) // 2
+        ids = tokenizer(text[:mid], add_special_tokens=False)["input_ids"]
+        if len(ids) == target_len:
+            target_ids = ids
+            break
+        if len(ids) < target_len:
+            lo = mid + 1
+        else:
+            hi = mid - 1
+    if target_ids is None:
+        ids = tokenizer(text[:lo], add_special_tokens=False)["input_ids"]
+        if len(ids) > target_len:
+            target_ids = ids[:target_len]
+        else:
+            filler = " data"
+            while len(ids) < target_len:
+                ids = tokenizer(tokenizer.decode(ids) + filler, add_special_tokens=False)["input_ids"]
+            target_ids = ids[:target_len]
+    prompt = tokenizer.decode(target_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)
+    # 断言精确长度
+    assert len(tokenizer(prompt, add_special_tokens=False)["input_ids"]) == target_len
+    return prompt
+# ========= V1 metrics 抽取工具 =========
+TTFT_METRIC_NAME = "vllm:time_to_first_token_seconds"
+TPOT_METRIC_NAME = "vllm:time_per_output_token_seconds"  # per-output-token latency
+def _iter_children_of_vector(vec_obj):
+    for attr in ("children", "metrics", "series", "values", "samples", "items"):
+        if hasattr(vec_obj, attr):
+            val = getattr(vec_obj, attr)
+            if isinstance(val, dict):
+                for v in val.values():
+                    yield v
+            else:
+                try:
+                    for v in val:
+                        yield v
+                except TypeError:
+                    pass
+def _collect_hist_sum_count(metrics, metric_name: str):
+    total_sum = 0.0
+    total_count = 0.0
+    for m in metrics:
+        mname = getattr(m, "name", None)
+        if mname != metric_name:
+            continue
+        # 直接 Histogram
+        if isinstance(m, Histogram) or m.__class__.__name__ == "Histogram":
+            total_sum += float(getattr(m, "sum", 0.0))
+            total_count += float(getattr(m, "count", 0.0))
+            continue
+        # Vector[Histogram]
+        if isinstance(m, Vector) or m.__class__.__name__ == "Vector":
+            for child in _iter_children_of_vector(m):
+                if isinstance(child, Histogram) or child.__class__.__name__ == "Histogram":
+                    total_sum += float(getattr(child, "sum", 0.0))
+                    total_count += float(getattr(child, "count", 0.0))
+    return total_sum, total_count
+def _metrics_snapshot(llm) -> Dict[str, float]:
+    try:
+        mets = llm.get_metrics()  # V1: 返回 Metric 列表（包含 Histogram/Vector 等）
+    except Exception:
+        return {"ttft_sum": 0.0, "ttft_cnt": 0.0, "tpot_sum": 0.0, "tpot_cnt": 0.0}
+    ttft_sum, ttft_cnt = _collect_hist_sum_count(mets, TTFT_METRIC_NAME)
+    tpot_sum, tpot_cnt = _collect_hist_sum_count(mets, TPOT_METRIC_NAME)
+    return {"ttft_sum": ttft_sum, "ttft_cnt": ttft_cnt, "tpot_sum": tpot_sum, "tpot_cnt": tpot_cnt}
+def _metrics_delta(before: dict, after: dict):
+    return {
+        "ttft_sum": after["ttft_sum"] - before["ttft_sum"],
+        "ttft_cnt": after["ttft_cnt"] - before["ttft_cnt"],
+        "tpot_sum": after["tpot_sum"] - before["tpot_sum"],
+        "tpot_cnt": after["tpot_cnt"] - before["tpot_cnt"],
+    }
+# ========= 带 NVTX 的 generate 包装 =========
+def decorated_generate(llm: LLM, prompts: List[str], params: SamplingParams):
+    return llm.generate(prompts, params)
+# ========= 统计格式化 =========
+def fmt_stats(x: List[float]) -> Tuple[float, float, float]:
+    xs = [v for v in x if (v == v)]  # 过滤 NaN
+    if not xs:
+        return (float("nan"), float("nan"), float("nan"))
+    return (statistics.mean(xs), statistics.median(xs), statistics.quantiles(xs, n=10)[-1])  # p90
+def main():
+    print("--- vLLM V1 基准测试（含 NVTX 标记）---")
+    print(f"模型: {MODEL_NAME}")
+    print(f"批量大小: {BATCH_SIZES}")
+    print(f"场景: {[s['name'] for s in SCENARIOS]}")
+    print("-" * 60)
+    if not torch.cuda.is_available():
+        print("错误：需要 CUDA GPU。")
+        return
+    print("加载分词器/模型中...")
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, use_fast=True, trust_remote_code=TRUST_REMOTE_CODE)
+    # 用 NVTX 标记模型加载阶段
+    nvtx.range_push("LLM_init")
+    llm = LLM(
+        model=MODEL_NAME,
+        tensor_parallel_size=TP,
+        dtype=DTYPE,
+        trust_remote_code=TRUST_REMOTE_CODE,
+        gpu_memory_utilization=GPU_MEM_UTIL,
+        max_num_seqs=1024,         # 足够覆盖本次扫描
+        max_model_len=4096,
+        disable_log_stats=False,  # 开启 V1 metrics 收集
+    )
+    nvtx.range_pop()
+    print("模型加载完成。")
+    for sc in SCENARIOS:
+        name = sc["name"]
+        prompt_tokens = sc["prompt_tokens"]
+        max_new_tokens = sc["max_new_tokens"]
+        print(f"\n===== 场景：{name} | prefill={prompt_tokens}, decode={max_new_tokens} =====")
+        # 准备精确长度 prompt
+        prompt_text = build_exact_token_prompt(tokenizer, prompt_tokens)
+        # 采样参数（贪心）
+        # sampling_params = SamplingParams(
+        #     max_tokens=max_new_tokens,
+        #     temperature=TEMPERATURE,
+        #     top_p=TOP_P,
+        #     seed=SEED,
+        #     n=1,
+        # )
+        sampling_params = SamplingParams(
+            max_tokens=max_new_tokens,      # 比如 512
+            # 关键点：
+            ignore_eos=True,          # 忽略 EOS，继续生成
+            stop=None,                # 不设 stop 字符串
+            stop_token_ids=[],        # 不设 stop token
+            # 选配：如果你的 vLLM 版本支持
+            min_tokens=max_new_tokens,      # 至少生成 N 个 token（≤ max_tokens）
+            temperature=0.0,
+            top_p=1.0,
+        )
+        # 记录每个 bs 的结果（便于后续统计或外部解析）
+        for bs in BATCH_SIZES:
+            print(f"\n--- 批量大小 bs={bs} ---")
+            prompts = [prompt_text] * bs
+            # 预热
+            # print("预热中...")
+            # nvtx.range_push(f"WARMUP [{name}] bs={bs}")
+            # _ = decorated_generate(llm, [prompts[0]], sampling_params)
+            # torch.cuda.synchronize()
+            # nvtx.range_pop()
+            # 正式计时与 V1 metrics
+            # nvtx.range_push(f"RUN [{name}] bs={bs}")
+            torch.cuda.synchronize()
+            snap_before = _metrics_snapshot(llm)
+            t0 = time.perf_counter()
+            nvtx.range_push(f"generate [{name}] bs={bs}")
+            outputs = decorated_generate(llm, prompts, sampling_params)
+            nvtx.range_pop()  # generate
+            torch.cuda.synchronize()
+            t1 = time.perf_counter()
+            snap_after = _metrics_snapshot(llm)
+            # nvtx.range_pop()  # RUN
+            duration = t1 - t0
+            # 统计 token 与吞吐
+            total_output_tokens = sum(len(o.outputs[0].token_ids) for o in outputs)
+            avg_prompt_tokens = sum(len(o.prompt_token_ids) for o in outputs) / bs
+            throughput = total_output_tokens / duration if duration > 0 else float("inf")
+            # 解析 V1 TTFT / 解码吞吐
+            delta = _metrics_delta(snap_before, snap_after)
+            if delta["ttft_cnt"] > 0:
+                ttft = delta["ttft_sum"] / delta["ttft_cnt"]
+            else:
+                ttft = float("nan")
+            if delta["tpot_cnt"] > 0:
+                avg_tpot = delta["tpot_sum"] / delta["tpot_cnt"]  # seconds/token
+                decode_tps = 1.0 / avg_tpot
+            else:
+                decode_tps = float("nan")
+            print(f"执行时间: {duration:.4f} s")
+            print(f"实际平均输入 tokens: {avg_prompt_tokens:.2f}（目标 {prompt_tokens}）")
+            print(f"生成总 tokens: {total_output_tokens}")
+            print(f"吞吐(生成tokens/秒): {throughput:.2f}")
+            print(f"TTFT (V1 metrics): {ttft:.4f} s")
+            print(f"解码吞吐 (V1 metrics): {decode_tps:.2f} tok/s")
+    print("\n完成。提示：在 Nsight Systems 中可通过 NVTX 区间快速定位各场景/批量的调用。")
+if __name__ == "__main__":
+    print(f"CUDA_VISIBLE_DEVICES = {os.getenv('CUDA_VISIBLE_DEVICES')}")
+    main()

traverse_bs_util_sim_prefill.log ADDED Viewed

	@@ -0,0 +1,382 @@

+WARNING: CPU IP/backtrace sampling not supported, disabling.
+Try the 'nsys status --environment' command to learn more.
+WARNING: CPU context switch tracing not supported, disabling.
+Try the 'nsys status --environment' command to learn more.
+INFO 08-13 19:12:40 [__init__.py:235] Automatically detected platform cuda.
+CUDA_VISIBLE_DEVICES = 3
+--- vLLM V1 基准测试（含 NVTX 标记）---
+模型: Qwen/Qwen2-1.5B
+批量大小: [1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024]
+场景: ['prefill640_decode1']
+------------------------------------------------------------
+加载分词器/模型中...
+INFO 08-13 19:12:50 [config.py:1604] Using max model len 4096
+INFO 08-13 19:12:50 [config.py:2434] Chunked prefill is enabled with max_num_batched_tokens=8192.
+INFO 08-13 19:12:55 [__init__.py:235] Automatically detected platform cuda.
+INFO 08-13 19:12:57 [core.py:572] Waiting for init message from front-end.
+INFO 08-13 19:12:57 [core.py:71] Initializing a V1 LLM engine (v0.10.0) with config: model='Qwen/Qwen2-1.5B', speculative_config=None, tokenizer='Qwen/Qwen2-1.5B', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, override_neuron_config={}, tokenizer_revision=None, trust_remote_code=True, dtype=torch.bfloat16, max_seq_len=4096, download_dir=None, load_format=LoadFormat.AUTO, tensor_parallel_size=1, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto,  device_config=cuda, decoding_config=DecodingConfig(backend='auto', disable_fallback=False, disable_any_whitespace=False, disable_additional_properties=False, reasoning_backend=''), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None), seed=0, served_model_name=Qwen/Qwen2-1.5B, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=True, chunked_prefill_enabled=True, use_async_output_proc=True, pooler_config=None, compilation_config={"level":3,"debug_dump_path":"","cache_dir":"","backend":"","custom_ops":[],"splitting_ops":["vllm.unified_attention","vllm.unified_attention_with_output","vllm.mamba_mixer2"],"use_inductor":true,"compile_sizes":[],"inductor_compile_config":{"enable_auto_functionalized_v2":false},"inductor_passes":{},"use_cudagraph":true,"cudagraph_num_of_warmups":1,"cudagraph_capture_sizes":[512,504,496,488,480,472,464,456,448,440,432,424,416,408,400,392,384,376,368,360,352,344,336,328,320,312,304,296,288,280,272,264,256,248,240,232,224,216,208,200,192,184,176,168,160,152,144,136,128,120,112,104,96,88,80,72,64,56,48,40,32,24,16,8,4,2,1],"cudagraph_copy_inputs":false,"full_cuda_graph":false,"max_capture_size":512,"local_cache_dir":null}
+INFO 08-13 19:12:59 [parallel_state.py:1102] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, TP rank 0, EP rank 0
+WARNING 08-13 19:12:59 [topk_topp_sampler.py:59] FlashInfer is not available. Falling back to the PyTorch-native implementation of top-p & top-k sampling. For the best performance, please install FlashInfer.
+INFO 08-13 19:12:59 [gpu_model_runner.py:1843] Starting to load model Qwen/Qwen2-1.5B...
+INFO 08-13 19:12:59 [gpu_model_runner.py:1875] Loading model from scratch...
+INFO 08-13 19:12:59 [cuda.py:290] Using Flash Attention backend on V1 engine.
+INFO 08-13 19:13:00 [weight_utils.py:296] Using model weights format ['*.safetensors']
+INFO 08-13 19:13:00 [weight_utils.py:349] No model.safetensors.index.json found in remote.
+INFO 08-13 19:13:01 [default_loader.py:262] Loading weights took 0.63 seconds
+INFO 08-13 19:13:01 [gpu_model_runner.py:1892] Model loading took 2.9105 GiB and 1.958554 seconds
+INFO 08-13 19:13:07 [backends.py:530] Using cache directory: /home/cy/.cache/vllm/torch_compile_cache/40b61c71e9/rank_0_0/backbone for vLLM's torch.compile
+INFO 08-13 19:13:07 [backends.py:541] Dynamo bytecode transform time: 5.46 s
+INFO 08-13 19:13:11 [backends.py:161] Directly load the compiled graph(s) for dynamic shape from the cache, took 3.856 s
+INFO 08-13 19:13:12 [monitor.py:34] torch.compile takes 5.46 s in total
+INFO 08-13 19:13:13 [gpu_worker.py:255] Available KV cache memory: 12.81 GiB
+INFO 08-13 19:13:13 [kv_cache_utils.py:833] GPU KV cache size: 479,536 tokens
+INFO 08-13 19:13:13 [kv_cache_utils.py:837] Maximum concurrency for 4,096 tokens per request: 117.07x
+INFO 08-13 19:13:15 [gpu_model_runner.py:2485] Graph capturing finished in 2 secs, took 0.49 GiB
+INFO 08-13 19:13:15 [core.py:193] init engine (profile, create kv cache, warmup model) took 14.10 seconds
+模型加载完成。
+===== 场景：prefill640_decode1 | prefill=640, decode=1 =====
+--- 批量大小 bs=1 ---
+执行时间: 0.0303 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 1
+吞吐(生成tokens/秒): 33.00
+TTFT (V1 metrics): 0.0268 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=2 ---
+执行时间: 0.0169 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 2
+吞吐(生成tokens/秒): 118.29
+TTFT (V1 metrics): 0.0123 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=4 ---
+执行时间: 0.0192 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 4
+吞吐(生成tokens/秒): 207.80
+TTFT (V1 metrics): 0.0145 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=8 ---
+执行时间: 0.0265 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 8
+吞吐(生成tokens/秒): 302.46
+TTFT (V1 metrics): 0.0151 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=16 ---
+执行时间: 0.0332 s
+实际平均输入 tokens: 640.00（目标 640）
+生成�� tokens: 16
+吞吐(生成tokens/秒): 481.67
+TTFT (V1 metrics): 0.0162 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=32 ---
+执行时间: 0.0553 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 32
+吞吐(生成tokens/秒): 578.82
+TTFT (V1 metrics): 0.0250 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=64 ---
+执行时间: 0.0975 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 64
+吞吐(生成tokens/秒): 656.64
+TTFT (V1 metrics): 0.0459 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=128 ---
+执行时间: 0.1830 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 128
+吞吐(生成tokens/秒): 699.40
+TTFT (V1 metrics): 0.0911 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=256 ---
+执行时间: 0.3573 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 256
+吞吐(生成tokens/秒): 716.54
+TTFT (V1 metrics): 0.1787 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=512 ---
+执行时间: 0.9187 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 512
+吞吐(生成tokens/秒): 557.31
+TTFT (V1 metrics): 0.3830 s
+解码吞吐 (V1 metrics): nan tok/s
+--- 批量大小 bs=1024 ---
+[rank0]:[W813 19:13:21.628248555 ProcessGroupNCCL.cpp:1479] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
+执行时间: 1.4298 s
+实际平均输入 tokens: 640.00（目标 640）
+生成总 tokens: 1024
+吞吐(生成tokens/秒): 716.16
+TTFT (V1 metrics): 0.7169 s
+解码吞吐 (V1 metrics): nan tok/s
+完成。提示：在 Nsight Systems 中可通过 NVTX 区间快速定位各场景/批量的调用。
+GPU 3: General Metrics for NVIDIA AD10x (any frequency)
+Generating '/tmp/nsys-report-be94.qdstrm'
+[3/8] Executing 'nvtx_sum' stats report
+ Time (%)  Total Time (ns)  Instances      Avg (ns)          Med (ns)         Min (ns)        Max (ns)     StdDev (ns)   Style                   Range
+ --------  ---------------  ---------  ----------------  ----------------  --------------  --------------  -----------  -------  --------------------------------------
+     91.5   34,255,644,442          1  34,255,644,442.0  34,255,644,442.0  34,255,644,442  34,255,644,442          0.0  PushPop  :LLM_init
+      3.8    1,429,765,717          1   1,429,765,717.0   1,429,765,717.0   1,429,765,717   1,429,765,717          0.0  PushPop  :generate [prefill640_decode1] bs=1024
+      2.5      918,598,349          1     918,598,349.0     918,598,349.0     918,598,349     918,598,349          0.0  PushPop  :generate [prefill640_decode1] bs=512
+      1.0      357,203,794          1     357,203,794.0     357,203,794.0     357,203,794     357,203,794          0.0  PushPop  :generate [prefill640_decode1] bs=256
+      0.5      182,944,803          1     182,944,803.0     182,944,803.0     182,944,803     182,944,803          0.0  PushPop  :generate [prefill640_decode1] bs=128
+      0.3       97,400,007          1      97,400,007.0      97,400,007.0      97,400,007      97,400,007          0.0  PushPop  :generate [prefill640_decode1] bs=64
+      0.1       55,233,548          1      55,233,548.0      55,233,548.0      55,233,548      55,233,548          0.0  PushPop  :generate [prefill640_decode1] bs=32
+      0.1       33,169,330          1      33,169,330.0      33,169,330.0      33,169,330      33,169,330          0.0  PushPop  :generate [prefill640_decode1] bs=16
+      0.1       30,210,666          1      30,210,666.0      30,210,666.0      30,210,666      30,210,666          0.0  PushPop  :generate [prefill640_decode1] bs=1
+      0.1       26,332,699          1      26,332,699.0      26,332,699.0      26,332,699      26,332,699          0.0  PushPop  :generate [prefill640_decode1] bs=8
+      0.1       19,203,429          1      19,203,429.0      19,203,429.0      19,203,429      19,203,429          0.0  PushPop  :generate [prefill640_decode1] bs=4
+      0.0       16,846,845          1      16,846,845.0      16,846,845.0      16,846,845      16,846,845          0.0  PushPop  :generate [prefill640_decode1] bs=2
+      0.0          105,272          2          52,636.0          52,636.0          49,581          55,691      4,320.4  PushPop  CCCL:cub::DeviceSegmentedRadixSort
+[4/8] Executing 'osrt_sum' stats report
+ Time (%)  Total Time (ns)  Num Calls      Avg (ns)        Med (ns)    Min (ns)      Max (ns)       StdDev (ns)              Name
+ --------  ---------------  ---------  ----------------  ------------  ---------  --------------  ----------------  ----------------------
+     31.6  297,381,519,943     14,617      20,344,908.0      34,539.0      1,030  22,258,440,024     414,761,021.5  pthread_cond_timedwait
+     26.2  246,799,144,239     18,218      13,546,994.4  10,061,235.5      1,012  25,048,066,633     321,112,693.8  epoll_wait
+     23.6  222,303,168,174         22  10,104,689,462.5   4,996,401.5     18,889  22,259,597,278  11,328,087,649.4  pthread_cond_wait
+      8.0   75,285,383,295      9,750       7,721,577.8       1,471.0      1,000  10,010,044,231     160,821,505.6  poll
+      6.2   58,364,790,020         28   2,084,456,786.4     130,001.0     11,516  10,000,121,902   3,904,360,688.6  sem_timedwait
+      3.7   34,631,283,784     31,001       1,117,102.2       2,501.0      1,000  20,354,884,016     126,413,489.1  read
+      0.5    4,262,251,691        375      11,366,004.5   6,769,937.0     25,915     606,781,491      45,247,323.7  sem_wait
+      0.0      435,782,688    193,508           2,252.0       1,290.0      1,001      78,395,275         178,359.2  munmap
+      0.0      295,324,731      8,442          34,982.8      10,400.0      1,002      28,343,070         412,455.7  ioctl
+      0.0      233,681,765        370         631,572.3       2,278.0      1,038      20,134,490       3,304,852.9  fopen
+      0.0      121,569,511         24       5,065,396.3   5,064,025.0  5,057,109       5,077,694           5,077.1  nanosleep
+      0.0       83,589,086     30,614           2,730.4       2,405.0      1,001          93,613           1,688.8  open64
+      0.0       82,741,435         44       1,880,487.2       3,387.5      1,008      76,213,344      11,471,905.8  waitpid
+      0.0       79,313,400         95         834,877.9       3,817.0      1,120      21,595,875       3,991,908.6  open
+      0.0       67,843,028         37       1,833,595.4     560,943.0      7,912      10,421,345       3,417,478.2  pthread_join
+      0.0       63,655,491         10       6,365,549.1      20,920.5      9,183      63,456,894      20,059,858.9  connect
+      0.0       49,611,881     12,926           3,838.1       2,497.5      1,000       1,578,326          21,496.3  mmap64
+      0.0       31,354,898     11,053           2,836.8       2,317.0      1,000          98,099           4,837.4  pthread_cond_signal
+      0.0       21,849,578        147         148,636.6      68,887.0     54,551      11,982,867         982,770.0  sleep
+      0.0       18,414,594      2,437           7,556.3       4,453.0      1,032       3,216,724          67,282.5  recv
+      0.0       18,204,857      1,636          11,127.7       8,142.5      1,000         828,819          26,164.8  pthread_mutex_lock
+      0.0       14,656,134      2,431           6,028.8       4,943.0      1,611          98,063           6,075.3  send
+      0.0       13,750,593      4,854           2,832.8       2,203.0      1,000          90,410           5,299.0  write
+      0.0        6,404,270        131          48,887.6      44,033.0     19,090         137,564          20,637.3  pthread_create
+      0.0        5,992,803        873           6,864.6       3,804.0      1,003          84,435           8,636.1  fgets
+      0.0        4,101,819      2,724           1,505.8       1,399.0      1,000          14,872             523.6  epoll_ctl
+      0.0        1,737,636        344           5,051.3       4,878.5      1,791          40,256           2,280.7  fopen64
+      0.0        1,580,944         62          25,499.1       3,739.0      1,061         361,517          63,617.6  futex
+      0.0        1,578,916         19          83,100.8      26,844.0     12,846         229,739          84,676.6  pthread_rwlock_wrlock
+      0.0        1,219,162      1,029           1,184.8       1,021.0      1,000           9,329             566.5  fclose
+      0.0          979,772        196           4,998.8       3,578.5      1,240          72,994           7,158.8  mmap
+      0.0          880,707          1         880,707.0     880,707.0    880,707         880,707               0.0  fork
+      0.0          817,580         10          81,758.0      50,097.5     12,485         189,119          64,533.4  pthread_rwlock_rdlock
+      0.0          344,919         65           5,306.4       4,286.0      1,901          13,820           2,831.0  pipe2
+      0.0          227,255         41           5,542.8       4,926.0      1,403          15,831           3,463.5  socket
+      0.0          176,218         19           9,274.6       2,447.0      1,013          63,892          16,667.1  bind
+      0.0          143,818         31           4,639.3       3,145.0      1,659          15,408           3,429.6  pthread_cond_broadcast
+      0.0           76,818          7          10,974.0      10,593.0      4,207          26,562           7,634.1  fread
+      0.0           58,634          5          11,726.8       9,913.0      6,172          20,449           5,556.1  accept4
+      0.0           44,580         27           1,651.1       1,250.0      1,001           4,508             919.9  fcntl
+      0.0           38,617         15           2,574.5       2,007.0      1,212           6,517           1,396.6  stat
+      0.0           38,220         22           1,737.3       1,771.5      1,040           2,400             357.0  sigaction
+      0.0           34,286         18           1,904.8       2,174.5      1,009           3,066             665.3  dup2
+      0.0           25,737          8           3,217.1       3,143.0      1,199           5,660           2,028.2  fflush
+      0.0           20,805          5           4,161.0       3,204.0      1,689           7,529           2,419.0  fwrite
+      0.0           19,991          4           4,997.8       4,954.0      4,665           5,418             314.8  lstat
+      0.0           17,854          8           2,231.8       1,745.0      1,320           3,550           1,017.9  pread
+      0.0           15,459         11           1,405.4       1,285.0      1,101           2,235             367.0  listen
+      0.0           14,610          4           3,652.5       3,564.5      2,768           4,713             801.6  flock
+      0.0           13,040          3           4,346.7       4,329.0      4,270           4,441              86.9  fputs_unlocked
+      0.0           11,981          5           2,396.2       2,771.0      1,202           2,927             711.6  mprotect
+      0.0            9,480          4           2,370.0       2,249.5      1,906           3,075             510.7  flockfile
+      0.0            6,404          1           6,404.0       6,404.0      6,404           6,404               0.0  kill
+      0.0            6,248          4           1,562.0       1,573.0      1,113           1,989             492.3  fstat
+      0.0            4,499          2           2,249.5       2,249.5      1,682           2,817             802.6  openat64
+      0.0            2,859          1           2,859.0       2,859.0      2,859           2,859               0.0  fputs
+      0.0            1,224          1           1,224.0       1,224.0      1,224           1,224               0.0  pthread_mutex_trylock
+[5/8] Executing 'cuda_api_sum' stats report
+ Time (%)  Total Time (ns)  Num Calls   Avg (ns)     Med (ns)    Min (ns)   Max (ns)    StdDev (ns)                     Name
+ --------  ---------------  ---------  -----------  -----------  --------  -----------  -----------  ------------------------------------------
+     34.3      812,140,032      3,083    263,425.2      6,428.0     2,928   98,188,542  1,916,618.2  cudaMemcpyAsync
+     21.8      516,021,139     36,947     13,966.5      6,360.0       775   58,987,857    445,775.0  cudaLaunchKernel
+      9.2      218,253,465      1,943    112,328.1     73,036.0    42,162    1,493,412    185,303.5  cudaGraphInstantiateWithFlags_v11040
+      8.5      199,893,902      1,066    187,517.7      3,358.5     1,531  143,716,638  4,421,443.3  cudaStreamSynchronize
+      7.6      180,431,780      2,135     84,511.4     31,189.0     5,473  114,045,672  2,467,576.1  cudaDeviceSynchronize
+      5.7      135,517,434      9,541     14,203.7     12,994.0     7,873       57,365      3,806.6  cudaGraphLaunch_v10000
+      4.4      103,077,576     32,794      3,143.2      3,734.0       625       42,966      2,040.8  cuLaunchKernel
+      2.3       53,907,221        222    242,825.3    116,370.0    60,442    2,394,882    358,819.3  cudaFree
+      1.7       40,503,891        348    116,390.5    108,995.5     4,264    1,127,684     59,276.4  cudaMalloc
+      1.1       25,697,655         10  2,569,765.5  2,671,025.0    56,954    4,429,810  1,454,311.0  cuLibraryLoadData
+      0.7       16,202,477      5,650      2,867.7      1,890.0       203      217,650      3,509.8  cudaMemsetAsync
+      0.4       10,179,971      9,641      1,055.9        400.0       260    4,173,119     44,067.9  cuKernelGetFunction
+      0.4       10,058,945        169     59,520.4     63,100.0    27,735      269,824     28,306.4  cuModuleLoadData
+      0.4       10,043,958     11,842        848.2        789.0       324       10,495        244.9  cudaStreamIsCapturing_v10000
+      0.4        9,094,302     18,895        481.3        447.0       292        6,998        140.7  cudaStreamGetCaptureInfo_v2_v11030
+      0.3        8,147,148      1,943      4,193.1      4,074.0     3,220       20,274        777.0  cudaStreamBeginCapture_v10000
+      0.3        7,560,536      1,943      3,891.2      3,854.0     2,420        9,751        520.1  cudaGraphDestroy_v10000
+      0.1        2,956,110        128     23,094.6      2,043.5     1,256    1,017,699    123,547.8  cudaStreamCreateWithPriority
+      0.1        2,497,642      1,943      1,285.5      1,271.0     1,001        1,848        114.9  cudaStreamEndCapture_v10000
+      0.1        1,589,818      1,943        818.2        750.0       631        2,645        247.1  cudaGraphGetNodes_v10000
+      0.1        1,331,824         14     95,130.3      5,133.0     3,547    1,267,757    337,507.7  cudaHostAlloc
+      0.0          224,235          8     28,029.4     27,965.5     8,873       65,303     19,334.0  cudaMemGetInfo
+      0.0          135,950        810        167.8        133.0        74        1,928        122.6  cuGetProcAddress_v2
+      0.0           16,525         19        869.7        347.0       273        4,275      1,102.2  cudaEventCreateWithFlags
+      0.0           13,697         15        913.1        798.0       354        2,245        499.4  cuLibraryGetKernel
+      0.0            8,006         14        571.9        561.0       298          931        170.8  cudaThreadExchangeStreamCaptureMode_v10010
+      0.0            5,200          1      5,200.0      5,200.0     5,200        5,200          0.0  cudaEventRecord
+      0.0            4,363          3      1,454.3      1,461.0     1,138        1,764        313.1  cuInit
+      0.0            3,658          1      3,658.0      3,658.0     3,658        3,658          0.0  cudaStreamWaitEvent
+      0.0            3,288          4        822.0        586.0        82        2,034        922.5  cuModuleGetLoadingMode
+      0.0            1,938          1      1,938.0      1,938.0     1,938        1,938          0.0  cudaEventDestroy
+      0.0            1,284          2        642.0        642.0       277        1,007        516.2  cudaGetDriverEntryPoint_v11030
+[6/8] Executing 'cuda_gpu_kern_sum' stats report
+ Time (%)  Total Time (ns)  Instances   Avg (ns)     Med (ns)    Min (ns)   Max (ns)   StdDev (ns)                                                  Name
+ --------  ---------------  ---------  -----------  -----------  ---------  ---------  -----------  ----------------------------------------------------------------------------------------------------
+     17.4      197,947,062      9,240     21,422.8     22,752.0      6,752     24,897      4,470.0  void flash::flash_fwd_splitkv_kernel<Flash_fwd_kernel_traits<(int)128, (int)64, (int)128, (int)4, (…
+     16.7      190,016,589      3,836     49,535.1     22,048.0      7,648    131,073     39,111.5  void cutlass::Kernel2<cutlass_80_tensorop_bf16_s16816gemm_relu_bf16_64x64_32x6_tn_align8>(T1::Param…
+     15.4      175,442,595        765    229,336.7     61,089.0     10,625    521,635    235,358.8  void cutlass::Kernel2<cutlass_80_wmma_tensorop_bf16_s161616gemm_bf16_16x16_128x2_tn_align8>(T1::Par…
+     13.9      157,634,573        924    170,600.2    162,625.5     40,224  1,415,047    225,630.4  ampere_bf16_s1688gemm_bf16_128x64_sliced1x2_ldg8_f2f_tn
+      6.9       78,474,832         28  2,802,672.6  2,803,582.0  2,787,886  2,808,622      5,629.9  ampere_bf16_s16816gemm_bf16_128x64_ldg8_f2f_tn
+      3.4       39,224,462          8  4,903,057.8  4,858,922.0  4,807,161  5,087,547    109,376.4  void at_cuda_detail::cub::DeviceSegmentedRadixSortKernel<at_cuda_detail::cub::DeviceRadixSortPolicy…
+      2.9       32,817,057        140    234,407.6    231,842.0    121,760    377,154     83,557.7  ampere_bf16_s1688gemm_bf16_64x128_sliced1x2_ldg8_f2f_tn
+      2.0       22,647,227        784     28,886.8     12,512.0     11,584     62,368     20,630.1  ampere_bf16_s16816gemm_bf16_64x64_ldg8_f2f_stages_64x5_tn
+      1.9       21,994,025      1,960     11,221.4      4,448.0      1,055    461,986     54,151.8  triton_poi_fused_mul_silu_1
+      1.8       20,374,153          4  5,093,538.3  5,082,874.0  4,907,865  5,300,540    193,781.3  void at_cuda_detail::cub::DeviceSegmentedRadixSortKernel<at_cuda_detail::cub::DeviceRadixSortPolicy…
+      1.7       19,801,839        924     21,430.6     21,408.0     21,024     21,856         89.0  void flash::flash_fwd_splitkv_kernel<Flash_fwd_kernel_traits<(int)128, (int)64, (int)128, (int)4, (…
+      1.3       14,291,848         28    510,423.1    511,794.0    469,890    513,858      7,978.2  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<signed char>, std::a…
+      1.2       13,732,511      9,240      1,486.2      1,472.0      1,120      2,176         99.3  void vllm::reshape_and_cache_flash_kernel<__nv_bfloat16, __nv_bfloat16, (vllm::Fp8KVCacheDataType)0…
+      1.2       13,254,405        142     93,340.9     51,632.5     50,176  3,030,607    350,548.4  ampere_bf16_s1688gemm_bf16_128x128_ldg8_f2f_stages_32x1_tn
+      0.9       10,247,270      1,960      5,228.2      3,488.0      1,536    110,881     12,591.6  triton_red_fused__to_copy_add_mean_mul_pow_rsqrt_2
+      0.9        9,752,980          4  2,438,245.0  2,430,621.0  2,388,877  2,502,861     56,361.3  void at::native::<unnamed>::cunn_SoftMaxForward<(int)4, float, float, float, at::native::<unnamed>:…
+      0.8        9,147,313         28    326,689.8    326,721.5    324,257    330,242      1,117.1  ampere_bf16_s1688gemm_bf16_128x128_ldg8_relu_f2f_stages_32x1_tn
+      0.7        8,448,874         97     87,101.8      8,865.0      7,040    498,466    160,499.7  std::enable_if<!T7, void>::type internal::gemvx::kernel<int, int, __nv_bfloat16, __nv_bfloat16, __n…
+      0.7        8,400,518        224     37,502.3     37,472.5     36,384     39,168        466.0  void cutlass::Kernel2<cutlass_80_tensorop_bf16_s16816gemm_relu_bf16_64x128_32x6_tn_align8>(T1::Para…
+      0.7        7,806,345          2  3,903,172.5  3,903,172.5  3,707,123  4,099,222    277,255.9  void at::native::_scatter_gather_elementwise_kernel<(int)128, (int)8, void at::native::_cuda_scatte…
+      0.6        7,228,323        336     21,512.9     21,440.0     21,088     22,624        351.4  ampere_bf16_s16816gemm_bf16_128x64_ldg8_relu_f2f_stages_64x3_tn
+      0.6        6,387,562      1,960      3,259.0      2,048.0      1,535     79,712      9,056.1  triton_red_fused__to_copy_add_mean_mul_pow_rsqrt_0
+      0.5        6,106,951        476     12,829.7     12,800.0     11,712     14,464        734.2  ampere_bf16_s16816gemm_bf16_64x64_ldg8_relu_f2f_stages_64x5_tn
+      0.5        5,899,904          4  1,474,976.0  1,474,552.0  1,473,800  1,477,000      1,402.5  void at::native::vectorized_elementwise_kernel<(int)4, at::native::<unnamed>::masked_fill_kernel(at…
+      0.4        4,854,373        334     14,534.1      5,823.0      5,376    714,116     76,872.2  void at::native::reduce_kernel<(int)512, (int)1, at::native::ReduceOp<float, at::native::ArgMaxOps<…
+      0.4        4,379,675        140     31,283.4     27,169.0     26,176     49,952      8,312.6  ampere_bf16_s1688gemm_bf16_128x64_sliced1x2_ldg8_relu_f2f_tn
+      0.4        4,089,246      1,890      2,163.6      1,824.0      1,344     22,400      2,259.9  triton_poi_fused_cat_3
+      0.4        3,995,829          2  1,997,914.5  1,997,914.5  1,996,875  1,998,954      1,470.1  void at::native::vectorized_elementwise_kernel<(int)4, at::native::BinaryFunctor<float, float, floa…
+      0.3        3,591,759         56     64,138.6     64,128.5     63,040     65,344        470.6  void cutlass::Kernel2<cutlass_80_wmma_tensorop_bf16_s161616gemm_bf16_32x32_64x1_tn_align8>(T1::Para…
+      0.3        3,536,964        332     10,653.5      4,672.0      2,143  1,004,742     77,469.7  void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIterator…
+      0.3        3,467,234      4,221        821.4        800.0        767      1,280         73.0  void at::native::vectorized_elementwise_kernel<(int)2, at::native::FillFunctor<long>, std::array<ch…
+      0.3        3,436,144          4    859,036.0    858,500.0    856,164    862,980      3,278.8  void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.3        3,187,952          2  1,593,976.0  1,593,976.0  1,571,272  1,616,680     32,108.3  void at::native::tensor_kernel_scan_innermost_dim<float, std::plus<float>>(T1 *, const T1 *, unsign…
+      0.3        2,855,430      1,512      1,888.5      1,697.0      1,312      2,880        453.6  void cublasLt::splitKreduce_kernel<(int)32, (int)16, int, __nv_bfloat16, __nv_bfloat16, float, (boo…
+      0.2        2,699,167      1,890      1,428.1      1,152.0        863     16,609      1,775.2  triton_poi_fused_view_5
+      0.2        2,650,106      1,890      1,402.2      1,344.0      1,215      6,336        546.8  triton_poi_fused_cat_4
+      0.2        2,583,886          2  1,291,943.0  1,291,943.0  1,291,463  1,292,423        678.8  void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.2        2,580,524          2  1,290,262.0  1,290,262.0  1,289,350  1,291,174      1,289.8  at::native::<unnamed>::fill_reverse_indices_kernel(long *, int, at::cuda::detail::IntDivider<unsign…
+      0.2        2,426,348        112     21,663.8     21,568.5      9,504     34,784     12,052.3  void cutlass::Kernel2<cutlass_80_wmma_tensorop_bf16_s161616gemm_bf16_32x32_128x2_tn_align8>(T1::Par…
+      0.1        1,501,800        466      3,222.7      3,040.0      2,880      7,456        428.1  void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void …
+      0.1        1,396,888        924      1,511.8      1,504.0      1,440      1,600         30.6  void vllm::merge_attn_states_kernel<__nv_bfloat16, (unsigned int)128>(T1 *, float *, const T1 *, co…
+      0.1        1,367,879          2    683,939.5    683,939.5    676,036    691,843     11,177.2  void at::native::<unnamed>::distribution_elementwise_grid_stride_kernel<float, (int)4, void at::nat…
+      0.1        1,147,814         28     40,993.4     40,864.0     40,352     42,560        490.6  ampere_bf16_s1688gemm_bf16_64x64_sliced1x4_ldg8_f2f_tn
+      0.1          957,606         28     34,200.2     34,768.5     17,600     35,200      3,259.5  std::enable_if<!T7, void>::type internal::gemvx::kernel<int, int, __nv_bfloat16, float, float, floa…
+      0.1          657,092         28     23,467.6     23,376.0     23,072     24,353        360.0  ampere_bf16_s16816gemm_bf16_128x64_ldg8_f2f_stages_32x6_tn
+      0.0          506,203        330      1,533.9      1,568.0      1,184      1,760        118.3  void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIterator…
+      0.0          417,887        332      1,258.7      1,248.0      1,120      1,695         43.2  void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIterator…
+      0.0          300,134        330        909.5        896.0        895        992         22.4  void at::native::unrolled_elementwise_kernel<at::native::CUDAFunctorOnSelf_add<int>, std::array<cha…
+      0.0          295,461        168      1,758.7      1,760.0      1,536      1,953        115.4  void cublasLt::splitKreduce_kernel<(int)32, (int)16, int, __nv_bfloat16, __nv_bfloat16, float, (boo…
+      0.0          290,763        330        881.1        865.0        863        960         23.0  void at::native::unrolled_elementwise_kernel<at::native::FillFunctor<int>, std::array<char *, (unsi…
+      0.0          257,951         70      3,685.0      3,040.0      2,048     35,776      3,957.7  triton_red_fused__to_copy_add_embedding_mean_mul_pow_rsqrt_0
+      0.0          230,204        258        892.3        896.0        863        960         21.7  void at::native::vectorized_elementwise_kernel<(int)2, at::native::FillFunctor<int>, std::array<cha…
+      0.0          191,518         70      2,736.0      2,112.0      1,631     40,096      4,544.2  triton_poi_fused_cat_1
+      0.0          157,536          1    157,536.0    157,536.0    157,536    157,536          0.0  void at::native::<unnamed>::CatArrayBatchedCopy_aligned16_contig<at::native::<unnamed>::OpaqueType<…
+      0.0          100,829         70      1,440.4      1,344.0      1,216      8,768        891.7  triton_poi_fused_cat_2
+      0.0           99,231         70      1,417.6      1,184.0        863     13,793      1,532.3  triton_poi_fused_view_3
+      0.0           80,032          1     80,032.0     80,032.0     80,032     80,032          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::bfloat16_copy_kernel_cuda(at::Te…
+      0.0           68,035         76        895.2        865.0        863      1,824        110.8  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<int>, std::array<cha…
+      0.0           64,195         58      1,106.8        896.0        864     11,360      1,372.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<c10::BFloat16>, std:…
+      0.0           43,840          1     43,840.0     43,840.0     43,840     43,840          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::sin_kernel_cuda(at::TensorIterat…
+      0.0           36,803         28      1,314.4      1,312.0      1,280      1,344         12.1  void cublasLt::splitKreduce_kernel<(int)32, (int)16, int, float, __nv_bfloat16, float, (bool)0, __n…
+      0.0           26,624          1     26,624.0     26,624.0     26,624     26,624          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::cos_kernel_cuda(at::TensorIterat…
+      0.0           19,361          1     19,361.0     19,361.0     19,361     19,361          0.0  void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.0           12,031         11      1,093.7        896.0        863      1,568        276.4  void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<float>, std::array<c…
+      0.0           10,815          2      5,407.5      5,407.5      5,151      5,664        362.7  void at::native::_scatter_gather_elementwise_kernel<(int)128, (int)8, void at::native::_cuda_scatte…
+      0.0            9,184          2      4,592.0      4,592.0      4,512      4,672        113.1  void at::native::<unnamed>::distribution_elementwise_grid_stride_kernel<float, (int)4, void at::nat…
+      0.0            3,616          2      1,808.0      1,808.0      1,600      2,016        294.2  void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::n…
+      0.0            3,521          2      1,760.5      1,760.5      1,696      1,825         91.2  void at::native::vectorized_elementwise_kernel<(int)2, at::native::CUDAFunctorOnOther_add<long>, st…
+      0.0            3,135          2      1,567.5      1,567.5      1,503      1,632         91.2  void at::native::vectorized_elementwise_kernel<(int)2, at::native::<unnamed>::where_kernel_impl(at:…
+      0.0            2,977          2      1,488.5      1,488.5      1,344      1,633        204.4  void at::native::vectorized_elementwise_kernel<(int)4, void at::native::compare_scalar_kernel<float…
+      0.0            2,976          2      1,488.0      1,488.0        992      1,984        701.4  void <unnamed>::elementwise_kernel_with_index<int, at::native::arange_cuda_out(const c10::Scalar &,…
+      0.0            2,912          2      1,456.0      1,456.0      1,344      1,568        158.4  void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnOther_add<float>, s…
+      0.0            2,240          1      2,240.0      2,240.0      2,240      2,240          0.0  void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::…
+      0.0            1,216          1      1,216.0      1,216.0      1,216      1,216          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::reciprocal_kernel_cuda(at::Tenso…
+      0.0            1,024          1      1,024.0      1,024.0      1,024      1,024          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<float, float, floa…
+      0.0              992          1        992.0        992.0        992        992          0.0  void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<float, float, floa…
+      0.0              896          1        896.0        896.0        896        896          0.0  void at::native::vectorized_elementwise_kernel<(int)2, at::native::FillFunctor<double>, std::array<…
+[7/8] Executing 'cuda_gpu_mem_time_sum' stats report
+ Time (%)  Total Time (ns)  Count   Avg (ns)     Med (ns)    Min (ns)    Max (ns)   StdDev (ns)            Operation
+ --------  ---------------  -----  -----------  -----------  ---------  ----------  -----------  ------------------------------
+     98.5      483,138,950  2,749    175,750.8        352.0        287  97,773,423  1,996,584.9  [CUDA memcpy Host-to-Device]
+      1.1        5,442,908      4  1,360,727.0  1,361,159.0  1,357,511   1,363,079      2,619.2  [CUDA memcpy Device-to-Device]
+      0.3        1,591,863  2,738        581.4        448.0        288       2,112        247.2  [CUDA memset]
+      0.1          374,720    330      1,135.5      1,120.0        864       1,760         85.7  [CUDA memcpy Device-to-Host]
+[8/8] Executing 'cuda_gpu_mem_size_sum' stats report
+ Total (MB)  Count  Avg (MB)  Med (MB)  Min (MB)  Max (MB)  StdDev (MB)            Operation
+ ----------  -----  --------  --------  --------  --------  -----------  ------------------------------
+  3,090.495  2,749     1.124     0.001     0.000   466.747       10.082  [CUDA memcpy Host-to-Device]
+  2,489.319      4   622.330   622.330   622.330   622.330        0.000  [CUDA memcpy Device-to-Device]
+      2.029  2,738     0.001     0.001     0.000     0.006        0.001  [CUDA memset]
+      0.008    330     0.000     0.000     0.000     0.000        0.000  [CUDA memcpy Device-to-Host]
+Generated:
+    /data/cy/kv_cache_vs_util/sim_traverse_bs/traverse_bs_util_sim_prefill.nsys-rep
+    /data/cy/kv_cache_vs_util/sim_traverse_bs/traverse_bs_util_sim_prefill.sqlite

traverse_bs_util_sim_prefill.nsys-rep ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:0b72e5b53980829e09ff4dfb58d9281aff1d6221df1e88e8ca3cf7e1976c1ee9
+size 17266751

traverse_bs_util_sim_prefill.py ADDED Viewed

	@@ -0,0 +1,262 @@

+import os
+import time
+import statistics
+from typing import List, Tuple, Dict
+import torch
+import torch.cuda.nvtx as nvtx
+from vllm import LLM, SamplingParams
+from transformers import AutoTokenizer
+# ========= 强制使用 vLLM V1 =========
+os.environ.setdefault("VLLM_USE_V1", "1")
+os.environ.setdefault("VLLM_WORKER_MULTIPROC_METHOD", "spawn")
+# 可选：打开 V1 metrics 统计
+os.environ.setdefault("VLLM_LOGGING_LEVEL", "INFO")
+# ========= 试图导入 V1 metrics 类型（兼容不同版本）=========
+try:
+    from vllm.v1.metrics.reader import Counter, Gauge, Histogram, Vector  # type: ignore
+except Exception:
+    Counter = Gauge = Histogram = Vector = type("X", (), {})  # dummy
+# ========= 配置 =========
+MODEL_NAME = "Qwen/Qwen2-1.5B"
+DTYPE = "bfloat16"
+TP = 1
+GPU_MEM_UTIL = 0.90
+TRUST_REMOTE_CODE = True
+# 场景：prefill=输入tokens，decode=输出tokens
+SCENARIOS = [
+    {"name": "prefill640_decode1", "prompt_tokens": 640, "max_new_tokens": 1},
+    # {"name": "prefill1_decode512", "prompt_tokens": 1,   "max_new_tokens": 512},
+    # {"name": "prefill640_decode512", "prompt_tokens": 640, "max_new_tokens": 512},
+]
+BATCH_SIZES = [1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024]
+SEED = 1234
+TEMPERATURE = 0.0
+TOP_P = 1.0
+WARMUP_PER_BS = 1  # 每个批次做一次预热
+# ========= 构造“精确 token 数量”的 prompt =========
+def build_exact_token_prompt(tokenizer, target_len: int) -> str:
+    if target_len <= 1:
+        # 最小化 prompt：用一个简单 token（避免空串导致0 token）
+        ids = tokenizer("A", add_special_tokens=False)["input_ids"]
+        if len(ids) >= 1:
+            return tokenizer.decode(ids[:1], skip_special_tokens=True, clean_up_tokenization_spaces=False)
+    base_text = (
+        "You are a helpful assistant. "
+        "Please analyze the following input and respond succinctly. "
+    )
+    chunk = " ".join(["data"] * 100) + ". "
+    text = base_text + chunk * 200  # 足够长的文本
+    lo, hi = 0, len(text)
+    target_ids = None
+    while lo <= hi:
+        mid = (lo + hi) // 2
+        ids = tokenizer(text[:mid], add_special_tokens=False)["input_ids"]
+        if len(ids) == target_len:
+            target_ids = ids
+            break
+        if len(ids) < target_len:
+            lo = mid + 1
+        else:
+            hi = mid - 1
+    if target_ids is None:
+        ids = tokenizer(text[:lo], add_special_tokens=False)["input_ids"]
+        if len(ids) > target_len:
+            target_ids = ids[:target_len]
+        else:
+            filler = " data"
+            while len(ids) < target_len:
+                ids = tokenizer(tokenizer.decode(ids) + filler, add_special_tokens=False)["input_ids"]
+            target_ids = ids[:target_len]
+    prompt = tokenizer.decode(target_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)
+    # 断言精确长度
+    assert len(tokenizer(prompt, add_special_tokens=False)["input_ids"]) == target_len
+    return prompt
+# ========= V1 metrics 抽取工具 =========
+TTFT_METRIC_NAME = "vllm:time_to_first_token_seconds"
+TPOT_METRIC_NAME = "vllm:time_per_output_token_seconds"  # per-output-token latency
+def _iter_children_of_vector(vec_obj):
+    for attr in ("children", "metrics", "series", "values", "samples", "items"):
+        if hasattr(vec_obj, attr):
+            val = getattr(vec_obj, attr)
+            if isinstance(val, dict):
+                for v in val.values():
+                    yield v
+            else:
+                try:
+                    for v in val:
+                        yield v
+                except TypeError:
+                    pass
+def _collect_hist_sum_count(metrics, metric_name: str):
+    total_sum = 0.0
+    total_count = 0.0
+    for m in metrics:
+        mname = getattr(m, "name", None)
+        if mname != metric_name:
+            continue
+        # 直接 Histogram
+        if isinstance(m, Histogram) or m.__class__.__name__ == "Histogram":
+            total_sum += float(getattr(m, "sum", 0.0))
+            total_count += float(getattr(m, "count", 0.0))
+            continue
+        # Vector[Histogram]
+        if isinstance(m, Vector) or m.__class__.__name__ == "Vector":
+            for child in _iter_children_of_vector(m):
+                if isinstance(child, Histogram) or child.__class__.__name__ == "Histogram":
+                    total_sum += float(getattr(child, "sum", 0.0))
+                    total_count += float(getattr(child, "count", 0.0))
+    return total_sum, total_count
+def _metrics_snapshot(llm) -> Dict[str, float]:
+    try:
+        mets = llm.get_metrics()  # V1: 返回 Metric 列表（包含 Histogram/Vector 等）
+    except Exception:
+        return {"ttft_sum": 0.0, "ttft_cnt": 0.0, "tpot_sum": 0.0, "tpot_cnt": 0.0}
+    ttft_sum, ttft_cnt = _collect_hist_sum_count(mets, TTFT_METRIC_NAME)
+    tpot_sum, tpot_cnt = _collect_hist_sum_count(mets, TPOT_METRIC_NAME)
+    return {"ttft_sum": ttft_sum, "ttft_cnt": ttft_cnt, "tpot_sum": tpot_sum, "tpot_cnt": tpot_cnt}
+def _metrics_delta(before: dict, after: dict):
+    return {
+        "ttft_sum": after["ttft_sum"] - before["ttft_sum"],
+        "ttft_cnt": after["ttft_cnt"] - before["ttft_cnt"],
+        "tpot_sum": after["tpot_sum"] - before["tpot_sum"],
+        "tpot_cnt": after["tpot_cnt"] - before["tpot_cnt"],
+    }
+# ========= 带 NVTX 的 generate 包装 =========
+def decorated_generate(llm: LLM, prompts: List[str], params: SamplingParams):
+    return llm.generate(prompts, params)
+# ========= 统计格式化 =========
+def fmt_stats(x: List[float]) -> Tuple[float, float, float]:
+    xs = [v for v in x if (v == v)]  # 过滤 NaN
+    if not xs:
+        return (float("nan"), float("nan"), float("nan"))
+    return (statistics.mean(xs), statistics.median(xs), statistics.quantiles(xs, n=10)[-1])  # p90
+def main():
+    print("--- vLLM V1 基准测试（含 NVTX 标记）---")
+    print(f"模型: {MODEL_NAME}")
+    print(f"批量大小: {BATCH_SIZES}")
+    print(f"场景: {[s['name'] for s in SCENARIOS]}")
+    print("-" * 60)
+    if not torch.cuda.is_available():
+        print("错误：需要 CUDA GPU。")
+        return
+    print("加载分词器/模型中...")
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, use_fast=True, trust_remote_code=TRUST_REMOTE_CODE)
+    # 用 NVTX 标记模型加载阶段
+    nvtx.range_push("LLM_init")
+    llm = LLM(
+        model=MODEL_NAME,
+        tensor_parallel_size=TP,
+        dtype=DTYPE,
+        trust_remote_code=TRUST_REMOTE_CODE,
+        gpu_memory_utilization=GPU_MEM_UTIL,
+        max_num_seqs=1024,         # 足够覆盖本次扫描
+        max_model_len=4096,
+        disable_log_stats=False,  # 开启 V1 metrics 收集
+    )
+    nvtx.range_pop()
+    print("模型加载完成。")
+    for sc in SCENARIOS:
+        name = sc["name"]
+        prompt_tokens = sc["prompt_tokens"]
+        max_new_tokens = sc["max_new_tokens"]
+        print(f"\n===== 场景：{name} | prefill={prompt_tokens}, decode={max_new_tokens} =====")
+        # 准备精确长度 prompt
+        prompt_text = build_exact_token_prompt(tokenizer, prompt_tokens)
+        # 采样参数（贪心）
+        sampling_params = SamplingParams(
+            max_tokens=max_new_tokens,
+            temperature=TEMPERATURE,
+            top_p=TOP_P,
+            seed=SEED,
+            n=1,
+        )
+        # 记录每个 bs 的结果（便于后续统计或外部解析）
+        for bs in BATCH_SIZES:
+            print(f"\n--- 批量大小 bs={bs} ---")
+            prompts = [prompt_text] * bs
+            # 预热
+            # print("预热中...")
+            # nvtx.range_push(f"WARMUP [{name}] bs={bs}")
+            # _ = decorated_generate(llm, [prompts[0]], sampling_params)
+            # torch.cuda.synchronize()
+            # nvtx.range_pop()
+            # 正式计时与 V1 metrics
+            # nvtx.range_push(f"RUN [{name}] bs={bs}")
+            torch.cuda.synchronize()
+            snap_before = _metrics_snapshot(llm)
+            t0 = time.perf_counter()
+            nvtx.range_push(f"generate [{name}] bs={bs}")
+            outputs = decorated_generate(llm, prompts, sampling_params)
+            nvtx.range_pop()  # generate
+            torch.cuda.synchronize()
+            t1 = time.perf_counter()
+            snap_after = _metrics_snapshot(llm)
+            # nvtx.range_pop()  # RUN
+            duration = t1 - t0
+            # 统计 token 与吞吐
+            total_output_tokens = sum(len(o.outputs[0].token_ids) for o in outputs)
+            avg_prompt_tokens = sum(len(o.prompt_token_ids) for o in outputs) / bs
+            throughput = total_output_tokens / duration if duration > 0 else float("inf")
+            # 解析 V1 TTFT / 解码吞吐
+            delta = _metrics_delta(snap_before, snap_after)
+            if delta["ttft_cnt"] > 0:
+                ttft = delta["ttft_sum"] / delta["ttft_cnt"]
+            else:
+                ttft = float("nan")
+            if delta["tpot_cnt"] > 0:
+                avg_tpot = delta["tpot_sum"] / delta["tpot_cnt"]  # seconds/token
+                decode_tps = 1.0 / avg_tpot
+            else:
+                decode_tps = float("nan")
+            print(f"执行时间: {duration:.4f} s")
+            print(f"实际平均输入 tokens: {avg_prompt_tokens:.2f}（目标 {prompt_tokens}）")
+            print(f"生成总 tokens: {total_output_tokens}")
+            print(f"吞吐(生成tokens/秒): {throughput:.2f}")
+            print(f"TTFT (V1 metrics): {ttft:.4f} s")
+            print(f"解码吞吐 (V1 metrics): {decode_tps:.2f} tok/s")
+    print("\n完成。提示：在 Nsight Systems 中可通过 NVTX 区间快速定位各场景/批量的调用。")
+if __name__ == "__main__":
+    print(f"CUDA_VISIBLE_DEVICES = {os.getenv('CUDA_VISIBLE_DEVICES')}")
+    main()