feat(dsv4): add B200 DeepSeek-V4-Pro dynamo-trt 8k1k disagg throughput recipes / 新增 B200 DeepSeek-V4-Pro dynamo-trt 8k1k 分离式吞吐量配置 #2721
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,125 @@ | ||
| name: disagg-b200-8k1k-1p2d-dep8-dep8-b4-eplb384-mtp3 | ||
| model: | ||
| path: deepseek-v4-pro | ||
| container: dynamo-trtllm | ||
| precision: fp4 | ||
| dynamo: | ||
| install: true | ||
| wheel: "1.4.0.dev20260807" | ||
| request_plane: tcp | ||
| resources: | ||
| gpu_type: b200 | ||
| prefill_nodes: 1 | ||
| prefill_workers: 1 | ||
| gpus_per_prefill: 8 | ||
| decode_nodes: 2 | ||
| decode_workers: 2 | ||
| gpus_per_decode: 8 | ||
| gpus_per_node: 8 | ||
| backend: | ||
| type: trtllm | ||
| prefill_environment: | ||
| TLLM_LOG_LEVEL: INFO | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
| TRTLLM_WORKER_DISABLE_GC: '1' | ||
| TRTLLM_ENABLE_PDL: '1' | ||
| ENROOT_ALLOW_DEV: 'yes' | ||
| NCCL_GRAPH_MIXING_SUPPORT: '0' | ||
| MIMALLOC_PURGE_DELAY: '0' | ||
| PYTORCH_CUDA_ALLOC_CONF: expandable_segments:True | ||
| TLLM_PROFILE_LOG_RANKS: all | ||
| TLLM_SPEC_DECODE_FORCE_NUM_ACCEPTED_TOKENS: '1.82' | ||
| UCX_TLS: rc,cuda_ipc,cuda_copy,sm,self,tcp | ||
| NIXL_PLUGIN_DIR: /opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/plugins | ||
| decode_environment: | ||
| TLLM_LOG_LEVEL: INFO | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
| TRTLLM_WORKER_DISABLE_GC: '1' | ||
| TRTLLM_ENABLE_PDL: '1' | ||
| ENROOT_ALLOW_DEV: 'yes' | ||
| NCCL_GRAPH_MIXING_SUPPORT: '0' | ||
| MIMALLOC_PURGE_DELAY: '0' | ||
| TLLM_PROFILE_LOG_RANKS: all | ||
| TLLM_SPEC_DECODE_FORCE_NUM_ACCEPTED_TOKENS: '1.82' | ||
| UCX_TLS: rc,cuda_ipc,cuda_copy,sm,self,tcp | ||
| NIXL_PLUGIN_DIR: /opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/plugins | ||
| trtllm_config: | ||
| prefill: | ||
| cache_transceiver_config: | ||
| backend: NIXL | ||
| kv_transfer_timeout_ms: 600000 | ||
| transceiver_runtime: PYTHON | ||
| cuda_graph_config: | ||
| disable_overlap_scheduler: true | ||
| enable_attention_dp: true | ||
| kv_cache_config: | ||
| dtype: fp8 | ||
| enable_block_reuse: false | ||
| free_gpu_memory_fraction: 0.5 | ||
| tokens_per_block: 128 | ||
| max_batch_size: 2 | ||
| max_num_tokens: 8192 | ||
| max_seq_len: 8232 | ||
| moe_config: | ||
| backend: TRTLLM | ||
| load_balancer: /configs/moe_load_balancer_ctx_ep4_384.yaml | ||
| moe_expert_parallel_size: 8 | ||
| pipeline_parallel_size: 1 | ||
| print_iter_log: true | ||
| speculative_config: | ||
| decoding_type: MTP | ||
| max_draft_len: 3 | ||
| tensor_parallel_size: 8 | ||
| decode: | ||
| cache_transceiver_config: | ||
| backend: NIXL | ||
| kv_transfer_timeout_ms: 600000 | ||
| transceiver_runtime: PYTHON | ||
| cuda_graph_config: | ||
| batch_sizes: | ||
| - 1 | ||
| - 2 | ||
| - 4 | ||
| enable_padding: true | ||
| enable_attention_dp: true | ||
| enable_lm_head_tp_in_adp: true | ||
| kv_cache_config: | ||
| dtype: fp8 | ||
| enable_block_reuse: false | ||
| free_gpu_memory_fraction: 0.8 | ||
| tokens_per_block: 128 | ||
| max_batch_size: 4 | ||
| max_num_tokens: 16 | ||
| max_seq_len: 9256 | ||
| moe_config: | ||
| backend: TRTLLM | ||
| load_balancer: /configs/moe_load_balancer_gen_ep8_slots384.yaml | ||
| use_low_precision_moe_combine: true | ||
| moe_expert_parallel_size: 8 | ||
| num_postprocess_workers: 4 | ||
| pipeline_parallel_size: 1 | ||
| print_iter_log: true | ||
| speculative_config: | ||
| decoding_type: MTP | ||
| max_draft_len: 3 | ||
| stream_interval: 100 | ||
| tensor_parallel_size: 8 | ||
| benchmark: | ||
| type: sa-bench | ||
| isl: 8192 | ||
| osl: 1024 | ||
| concurrencies: "64" | ||
| req_rate: inf | ||
| random_range_ratio: 0.8 | ||
| use_chat_template: true | ||
| custom_tokenizer: "sa_bench_tokenizers.sglang_deepseek_v4.SGLangDeepseekV4Tokenizer" | ||
| frontend: | ||
| type: dynamo | ||
| enable_multiple_frontends: false | ||
| slurm: | ||
| time_limit: '3:00:00' | ||
| health_check: | ||
| max_attempts: 540 | ||
| interval_seconds: 10 | ||
| environment: | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,125 @@ | ||
| name: disagg-b200-8k1k-1p4d-dep8-dep8-b2-eplb384-mtp3 | ||
| model: | ||
| path: deepseek-v4-pro | ||
| container: dynamo-trtllm | ||
| precision: fp4 | ||
| dynamo: | ||
| install: true | ||
| wheel: "1.4.0.dev20260807" | ||
| request_plane: tcp | ||
| resources: | ||
| gpu_type: b200 | ||
| prefill_nodes: 1 | ||
| prefill_workers: 1 | ||
| gpus_per_prefill: 8 | ||
| decode_nodes: 4 | ||
| decode_workers: 4 | ||
| gpus_per_decode: 8 | ||
| gpus_per_node: 8 | ||
| backend: | ||
| type: trtllm | ||
| prefill_environment: | ||
| TLLM_LOG_LEVEL: INFO | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
| TRTLLM_WORKER_DISABLE_GC: '1' | ||
| TRTLLM_ENABLE_PDL: '1' | ||
| ENROOT_ALLOW_DEV: 'yes' | ||
| NCCL_GRAPH_MIXING_SUPPORT: '0' | ||
| MIMALLOC_PURGE_DELAY: '0' | ||
| PYTORCH_CUDA_ALLOC_CONF: expandable_segments:True | ||
| TLLM_PROFILE_LOG_RANKS: all | ||
| TLLM_SPEC_DECODE_FORCE_NUM_ACCEPTED_TOKENS: '1.82' | ||
| UCX_TLS: rc,cuda_ipc,cuda_copy,sm,self,tcp | ||
| NIXL_PLUGIN_DIR: /opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/plugins | ||
| decode_environment: | ||
| TLLM_LOG_LEVEL: INFO | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
| TRTLLM_WORKER_DISABLE_GC: '1' | ||
| TRTLLM_ENABLE_PDL: '1' | ||
| ENROOT_ALLOW_DEV: 'yes' | ||
| NCCL_GRAPH_MIXING_SUPPORT: '0' | ||
| MIMALLOC_PURGE_DELAY: '0' | ||
| TLLM_PROFILE_LOG_RANKS: all | ||
| TLLM_SPEC_DECODE_FORCE_NUM_ACCEPTED_TOKENS: '1.82' | ||
| UCX_TLS: rc,cuda_ipc,cuda_copy,sm,self,tcp | ||
| NIXL_PLUGIN_DIR: /opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/plugins | ||
| trtllm_config: | ||
| prefill: | ||
| cache_transceiver_config: | ||
| backend: NIXL | ||
| kv_transfer_timeout_ms: 600000 | ||
| transceiver_runtime: PYTHON | ||
| cuda_graph_config: | ||
| disable_overlap_scheduler: true | ||
| enable_attention_dp: true | ||
| kv_cache_config: | ||
| dtype: fp8 | ||
| enable_block_reuse: false | ||
| free_gpu_memory_fraction: 0.5 | ||
| tokens_per_block: 128 | ||
| max_batch_size: 2 | ||
| max_num_tokens: 8192 | ||
| max_seq_len: 8232 | ||
| moe_config: | ||
| backend: TRTLLM | ||
| load_balancer: /configs/moe_load_balancer_ctx_ep4_384.yaml | ||
| moe_expert_parallel_size: 8 | ||
| pipeline_parallel_size: 1 | ||
| print_iter_log: true | ||
| speculative_config: | ||
| decoding_type: MTP | ||
| max_draft_len: 3 | ||
| tensor_parallel_size: 8 | ||
| decode: | ||
| cache_transceiver_config: | ||
| backend: NIXL | ||
| kv_transfer_timeout_ms: 600000 | ||
| transceiver_runtime: PYTHON | ||
| cuda_graph_config: | ||
| batch_sizes: | ||
| - 1 | ||
| - 2 | ||
| - 4 | ||
| enable_padding: true | ||
| enable_attention_dp: true | ||
| enable_lm_head_tp_in_adp: true | ||
| kv_cache_config: | ||
| dtype: fp8 | ||
| enable_block_reuse: false | ||
| free_gpu_memory_fraction: 0.8 | ||
| tokens_per_block: 128 | ||
| max_batch_size: 2 | ||
|
There was a problem hiding this comment. CUDA graph batch exceeds max batchMedium Severity Decode   Reviewed by Cursor Bugbot for commit 14289a7. Configure here. |
||
| max_num_tokens: 8 | ||
| max_seq_len: 9256 | ||
| moe_config: | ||
| backend: TRTLLM | ||
| load_balancer: /configs/moe_load_balancer_gen_ep8_slots384.yaml | ||
| use_low_precision_moe_combine: true | ||
| moe_expert_parallel_size: 8 | ||
| num_postprocess_workers: 4 | ||
| pipeline_parallel_size: 1 | ||
| print_iter_log: true | ||
| speculative_config: | ||
| decoding_type: MTP | ||
| max_draft_len: 3 | ||
| stream_interval: 100 | ||
| tensor_parallel_size: 8 | ||
| benchmark: | ||
| type: sa-bench | ||
| isl: 8192 | ||
| osl: 1024 | ||
| concurrencies: "64" | ||
| req_rate: inf | ||
| random_range_ratio: 0.8 | ||
| use_chat_template: true | ||
| custom_tokenizer: "sa_bench_tokenizers.sglang_deepseek_v4.SGLangDeepseekV4Tokenizer" | ||
| frontend: | ||
| type: dynamo | ||
| enable_multiple_frontends: false | ||
| slurm: | ||
| time_limit: '3:00:00' | ||
| health_check: | ||
| max_attempts: 540 | ||
| interval_seconds: 10 | ||
| environment: | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,123 @@ | ||
| name: disagg-b200-8k1k-1p4d-dep8-tep8-b4-mtp3 | ||
| model: | ||
| path: deepseek-v4-pro | ||
| container: dynamo-trtllm | ||
| precision: fp4 | ||
| dynamo: | ||
| install: true | ||
| wheel: "1.4.0.dev20260807" | ||
| request_plane: tcp | ||
| resources: | ||
| gpu_type: b200 | ||
| prefill_nodes: 1 | ||
| prefill_workers: 1 | ||
| gpus_per_prefill: 8 | ||
| decode_nodes: 4 | ||
| decode_workers: 4 | ||
| gpus_per_decode: 8 | ||
| gpus_per_node: 8 | ||
| backend: | ||
| type: trtllm | ||
| prefill_environment: | ||
| TLLM_LOG_LEVEL: INFO | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
| TRTLLM_WORKER_DISABLE_GC: '1' | ||
| TRTLLM_ENABLE_PDL: '1' | ||
| ENROOT_ALLOW_DEV: 'yes' | ||
| NCCL_GRAPH_MIXING_SUPPORT: '0' | ||
| MIMALLOC_PURGE_DELAY: '0' | ||
| PYTORCH_CUDA_ALLOC_CONF: expandable_segments:True | ||
| TLLM_PROFILE_LOG_RANKS: all | ||
| TLLM_SPEC_DECODE_FORCE_NUM_ACCEPTED_TOKENS: '1.82' | ||
| UCX_TLS: rc,cuda_ipc,cuda_copy,sm,self,tcp | ||
| NIXL_PLUGIN_DIR: /opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/plugins | ||
| decode_environment: | ||
| TLLM_LOG_LEVEL: INFO | ||
| TRTLLM_SERVER_DISABLE_GC: '1' | ||
| TRTLLM_WORKER_DISABLE_GC: '1' | ||
| TRTLLM_ENABLE_PDL: '1' | ||
| ENROOT_ALLOW_DEV: 'yes' | ||
| NCCL_GRAPH_MIXING_SUPPORT: '0' | ||
| MIMALLOC_PURGE_DELAY: '0' | ||
| TLLM_PROFILE_LOG_RANKS: all | ||
| TLLM_SPEC_DECODE_FORCE_NUM_ACCEPTED_TOKENS: '1.82' | ||
| UCX_TLS: rc,cuda_ipc,cuda_copy,sm,self,tcp | ||
| NIXL_PLUGIN_DIR: /opt/nvidia/nvda_nixl/lib/x86_64-linux-gnu/plugins | ||
| trtllm_config: | ||
| prefill: | ||
| cache_transceiver_config: | ||
| backend: NIXL | ||
| kv_transfer_timeout_ms: 600000 | ||
| transceiver_runtime: PYTHON | ||
| cuda_graph_config: | ||
| disable_overlap_scheduler: true | ||
| enable_attention_dp: true | ||
| kv_cache_config: | ||
| dtype: fp8 | ||
| enable_block_reuse: false | ||
| free_gpu_memory_fraction: 0.5 | ||
| tokens_per_block: 128 | ||
| max_batch_size: 2 | ||
| max_num_tokens: 8192 | ||
| max_seq_len: 8232 | ||
| moe_config: | ||
| backend: TRTLLM | ||
| moe_expert_parallel_size: 8 | ||
| pipeline_parallel_size: 1 | ||
| print_iter_log: true | ||
| speculative_config: | ||
| decoding_type: MTP | ||
| max_draft_len: 3 | ||
| tensor_parallel_size: 8 | ||
| decode: | ||
| cache_transceiver_config: | ||
| backend: NIXL | ||
| kv_transfer_timeout_ms: 600000 | ||
| transceiver_runtime: PYTHON | ||
| cuda_graph_config: | ||
| batch_sizes: | ||
| - 1 | ||
| - 2 | ||
| - 4 | ||
| enable_padding: true | ||
| enable_attention_dp: false | ||
| enable_lm_head_tp_in_adp: false | ||
| kv_cache_config: | ||
| dtype: fp8 | ||
| enable_block_reuse: false | ||
| free_gpu_memory_fraction: 0.9 | ||
| tokens_per_block: 128 | ||
| max_batch_size: 4 | ||
| max_num_tokens: 16 | ||
| max_seq_len: 9256 | ||
| moe_config: | ||
| backend: TRTLLM | ||
| use_low_precision_moe_combine: true | ||
| moe_expert_parallel_size: 8 | ||
| num_postprocess_workers: 4 | ||
| pipeline_parallel_size: 1 | ||
| print_iter_log: true | ||
| speculative_config: | ||
| decoding_type: MTP | ||
| max_draft_len: 3 | ||
| stream_interval: 100 | ||
| tensor_parallel_size: 8 | ||
| benchmark: | ||
| type: sa-bench | ||
| isl: 8192 | ||
| osl: 1024 | ||
| concurrencies: "16" | ||
| req_rate: inf | ||
| random_range_ratio: 0.8 | ||
| use_chat_template: true | ||
| custom_tokenizer: "sa_bench_tokenizers.sglang_deepseek_v4.SGLangDeepseekV4Tokenizer" | ||
| frontend: | ||
| type: dynamo | ||
| enable_multiple_frontends: false | ||
| slurm: | ||
| time_limit: '3:00:00' | ||
| health_check: | ||
| max_attempts: 540 | ||
| interval_seconds: 10 | ||
| environment: | ||
| TRTLLM_SERVER_DISABLE_GC: '1' |
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Prefill EPLB config EP mismatch
Medium Severity
Prefill
moe_config.load_balancerpoints atmoe_load_balancer_ctx_ep4_384.yamlwhilemoe_expert_parallel_sizeis8. That EPLB table is built for EP4 slot sharding, so DEP8 prefill places experts with the wrong rank mapping and can unbalance MoE load or OOM a subset of ranks.Additional Locations (1)
benchmarks/multi_node/srt-slurm-recipes/trtllm/deepseek-v4/8k1k/b200/mtp/disagg-b200-8k1k-1p4d-dep8-dep8-b2-eplb384-mtp3.yaml#L64-L66Reviewed by Cursor Bugbot for commit 0aafb03. Configure here.