#!/bin/bash
# SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
set -e
trap 'echo Cleaning up...; kill 0' EXIT

MODEL="Qwen/Qwen3-0.6B"
HTTP_PORT="${DYN_HTTP_PORT:-8000}"
echo "=========================================="
echo "Launching Disaggregated + KVBM 2P+2D (4 GPUs)"
echo "=========================================="
echo "Model:       $MODEL"
echo "Frontend:    http://localhost:$HTTP_PORT"
echo "=========================================="
echo ""
echo "Example test command:"
echo ""
echo "  curl http://localhost:${HTTP_PORT}/v1/chat/completions \\"
echo "    -H 'Content-Type: application/json' \\"
echo "    -d '{"
echo "      \"model\": \"${MODEL}\","
echo "      \"messages\": [{\"role\": \"user\", \"content\": \"Explain why Roger Federer is considered one of the greatest tennis players of all time\"}],"
echo "      \"max_tokens\": 32"
echo "    }'"
echo ""
echo "=========================================="

# run ingress with KV router
# dynamo.frontend accepts either --http-port flag or DYN_HTTP_PORT env var (defaults to 8000)
python -m dynamo.frontend --router-mode kv &

# run decode workers on GPU 0 and 1, without enabling KVBM
# NOTE: remove --enforce-eager for production use
CUDA_VISIBLE_DEVICES=0 python3 -m dynamo.vllm --model "$MODEL" --kv-transfer-config '{"kv_connector":"NixlConnector","kv_role":"kv_both"}' --enforce-eager --disaggregation-mode decode &
VLLM_NIXL_SIDE_CHANNEL_PORT=20097 \
CUDA_VISIBLE_DEVICES=1 python3 -m dynamo.vllm --model "$MODEL" --kv-transfer-config '{"kv_connector":"NixlConnector","kv_role":"kv_both"}' --enforce-eager --disaggregation-mode decode &

# run prefill workers on GPU 2 and 3 with KVBM enabled using 20GB of CPU cache
# NOTE: use different barrier id prefixes for each prefill worker to avoid conflicts
# NOTE: remove --enforce-eager for production use
VLLM_NIXL_SIDE_CHANNEL_PORT=20098 \
DYN_KVBM_CPU_CACHE_GB=20 \
CUDA_VISIBLE_DEVICES=2 \
  python3 -m dynamo.vllm \
    --model "$MODEL" \
    --disaggregation-mode prefill \
    --kv-transfer-config '{"kv_connector":"PdConnector","kv_role":"kv_both","kv_connector_extra_config":{"connectors":[{"kv_connector":"DynamoConnector","kv_connector_module_path":"kvbm.vllm_integration.connector","kv_role":"kv_both"},{"kv_connector":"NixlConnector","kv_role":"kv_both"}]},"kv_connector_module_path":"kvbm.vllm_integration.connector"}' \
    --enforce-eager \
    --kv-events-config '{"publisher":"zmq","topic":"kv-events","endpoint":"tcp://*:20082","enable_kv_cache_events":true}' &

VLLM_NIXL_SIDE_CHANNEL_PORT=20099 \
DYN_KVBM_LEADER_ZMQ_PUB_PORT=56003 \
DYN_KVBM_LEADER_ZMQ_ACK_PORT=56004 \
DYN_KVBM_CPU_CACHE_GB=20 \
CUDA_VISIBLE_DEVICES=3 \
  python3 -m dynamo.vllm \
    --model "$MODEL" \
    --disaggregation-mode prefill \
    --kv-transfer-config '{"kv_connector":"PdConnector","kv_role":"kv_both","kv_connector_extra_config":{"connectors":[{"kv_connector":"DynamoConnector","kv_connector_module_path":"kvbm.vllm_integration.connector","kv_role":"kv_both"},{"kv_connector":"NixlConnector","kv_role":"kv_both"}]},"kv_connector_module_path":"kvbm.vllm_integration.connector"}' \
    --enforce-eager \
    --kv-events-config '{"publisher":"zmq","topic":"kv-events","endpoint":"tcp://*:20083","enable_kv_cache_events":true}'