Browse Source

fix: enhance comments in compose.yaml for clearer engine selection and GPU configuration guidance

myhloli 1 week ago
parent
commit
a149a8da50
1 changed files with 60 additions and 31 deletions
  1. 60 31
      docker/compose.yaml

+ 60 - 31
docker/compose.yaml

@@ -10,18 +10,29 @@ services:
       MINERU_MODEL_SOURCE: local
       MINERU_MODEL_SOURCE: local
     entrypoint: mineru-openai-server
     entrypoint: mineru-openai-server
     command:
     command:
-      # !!!The lmdeploy and vLLM engines cannot be enabled simultaneously.!!!
-      --engine vllm  # Choose between 'vllm' or 'lmdeploy' engine
-      # parameters for vllm-engine
+      # ==================== Engine Selection ====================
+      # WARNING: Only ONE engine can be enabled at a time!
+      # Choose 'vllm' OR 'lmdeploy' (uncomment one line below)
+      --engine vllm
+      # --engine lmdeploy
+      
+      # ==================== vLLM Engine Parameters ====================
+      # Uncomment if using --engine vllm
       --host 0.0.0.0
       --host 0.0.0.0
       --port 30000
       --port 30000
-      # --data-parallel-size 2  # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode
-      # --gpu-memory-utilization 0.5  # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter, if VRAM issues persist, try lowering it further to `0.4` or below.
-      # parameters for lmdeploy-engine
+      # Multi-GPU configuration (increase throughput)
+      # --data-parallel-size 2
+      # Single GPU memory optimization (reduce if VRAM insufficient)
+      # --gpu-memory-utilization 0.5  # Try 0.4 or lower if issues persist
+
+      # ==================== LMDeploy Engine Parameters ====================
+      # Uncomment if using --engine lmdeploy
       # --server-name 0.0.0.0
       # --server-name 0.0.0.0
-      # --server-port  30000
-      # --dp 2  # If using multiple GPUs, increase throughput using lmdeploy's multi-GPU parallel mode
-      # --cache-max-entry-count 0.5  # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter, if VRAM issues persist, try lowering it further to `0.4` or below.
+      # --server-port 30000
+      # Multi-GPU configuration (increase throughput)
+      # --dp 2
+      # Single GPU memory optimization (reduce if VRAM insufficient)
+      # --cache-max-entry-count 0.5  # Try 0.4 or lower if issues persist
     ulimits:
     ulimits:
       memlock: -1
       memlock: -1
       stack: 67108864
       stack: 67108864
@@ -33,7 +44,7 @@ services:
         reservations:
         reservations:
           devices:
           devices:
             - driver: nvidia
             - driver: nvidia
-              device_ids: ["0"]
+              device_ids: ["0"]  # Modify for multiple GPUs: ["0", "1"]
               capabilities: [gpu]
               capabilities: [gpu]
 
 
   mineru-api:
   mineru-api:
@@ -47,14 +58,21 @@ services:
       MINERU_MODEL_SOURCE: local
       MINERU_MODEL_SOURCE: local
     entrypoint: mineru-api
     entrypoint: mineru-api
     command:
     command:
+      # ==================== Server Configuration ====================
       --host 0.0.0.0
       --host 0.0.0.0
       --port 8000
       --port 8000
-      # parameters for vllm-engine
-      # --data-parallel-size 2  # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode
-      # --gpu-memory-utilization 0.5  # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter, if VRAM issues persist, try lowering it further to `0.4` or below.
-      # parameters for lmdeploy-engine
-      # --dp 2  # If using multiple GPUs, increase throughput using lmdeploy's multi-GPU parallel mode
-      # --cache-max-entry-count 0.5  # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter, if VRAM issues persist, try lowering it further to `0.4` or below.
+
+      # ==================== vLLM Engine Parameters ====================
+      # Multi-GPU configuration
+      # --data-parallel-size 2
+      # Single GPU memory optimization
+      # --gpu-memory-utilization 0.5  # Try 0.4 or lower if VRAM insufficient
+
+      # ==================== LMDeploy Engine Parameters ====================
+      # Multi-GPU configuration
+      # --dp 2
+      # Single GPU memory optimization
+      # --cache-max-entry-count 0.5  # Try 0.4 or lower if VRAM insufficient
     ulimits:
     ulimits:
       memlock: -1
       memlock: -1
       stack: 67108864
       stack: 67108864
@@ -64,8 +82,8 @@ services:
         reservations:
         reservations:
           devices:
           devices:
             - driver: nvidia
             - driver: nvidia
-              device_ids: [ "0" ]
-              capabilities: [ gpu ]
+              device_ids: ["0"]  # Modify for multiple GPUs: ["0", "1"]
+              capabilities: [gpu]
 
 
   mineru-gradio:
   mineru-gradio:
     image: mineru:latest
     image: mineru:latest
@@ -78,19 +96,30 @@ services:
       MINERU_MODEL_SOURCE: local
       MINERU_MODEL_SOURCE: local
     entrypoint: mineru-gradio
     entrypoint: mineru-gradio
     command:
     command:
+      # ==================== Gradio Server Configuration ====================
       --server-name 0.0.0.0
       --server-name 0.0.0.0
       --server-port 7860
       --server-port 7860
-      # --enable-api false  # If you want to disable the API, set this to false
-      # --max-convert-pages 20  # If you want to limit the number of pages for conversion, set this to a specific number
-      # parameters for vllm-engine
-      --enable-vllm-engine true  # Enable the vllm engine for Gradio
-      # --data-parallel-size 2  # If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode
-      # --gpu-memory-utilization 0.5  # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter, if VRAM issues persist, try lowering it further to `0.4` or below.
-      # parameters for lmdeploy-engine
-      # !!!The lmdeploy and vLLM engines cannot be enabled simultaneously. Please ensure that at most only one engine is active at any given time.!!!
-      # --enable-lmdeploy-engine true  # Enable the lmdeploy engine for Gradio
-      # --dp 2  # If using multiple GPUs, increase throughput using lmdeploy's multi-GPU parallel mode
-      # --cache-max-entry-count 0.5  # If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter, if VRAM issues persist, try lowering it further to `0.4` or below.
+      
+      # ==================== Gradio Feature Settings ====================
+      # --enable-api false  # Disable API endpoint
+      # --max-convert-pages 20  # Limit conversion page count
+      
+      # ==================== Engine Selection ====================
+      # WARNING: Only ONE engine can be enabled at a time!
+      
+      # Option 1: vLLM Engine (recommended for most users)
+      --enable-vllm-engine true
+      # Multi-GPU configuration
+      # --data-parallel-size 2
+      # Single GPU memory optimization
+      # --gpu-memory-utilization 0.5  # Try 0.4 or lower if VRAM insufficient
+
+      # Option 2: LMDeploy Engine
+      # --enable-lmdeploy-engine true
+      # Multi-GPU configuration
+      # --dp 2
+      # Single GPU memory optimization
+      # --cache-max-entry-count 0.5  # Try 0.4 or lower if VRAM insufficient
     ulimits:
     ulimits:
       memlock: -1
       memlock: -1
       stack: 67108864
       stack: 67108864
@@ -100,5 +129,5 @@ services:
         reservations:
         reservations:
           devices:
           devices:
             - driver: nvidia
             - driver: nvidia
-              device_ids: [ "0" ]
-              capabilities: [ gpu ]
+              device_ids: ["0"]  # Modify for multiple GPUs: ["0", "1"]
+              capabilities: [gpu]