{
  "uid": "qwen-l20-evidence-cache",
  "title": "Qwen L20 · 在线缓存路由交叉测试",
  "schemaVersion": 39,
  "version": 1,
  "tags": [
    "AIBrix",
    "Qwen3.8",
    "L20",
    "light"
  ],
  "style": "light",
  "timezone": "Asia/Shanghai",
  "refresh": "",
  "time": {
    "from": "now-1h",
    "to": "now"
  },
  "editable": true,
  "panels": [
    {
      "id": 2,
      "title": "引擎 TTFT P95",
      "type": "timeseries",
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "gridPos": {
        "x": 0,
        "y": 0,
        "w": 12,
        "h": 9
      },
      "targets": [
        {
          "refId": "A",
          "expr": "histogram_quantile(0.95,sum by(le,node_alias)(rate(vllm:time_to_first_token_seconds_bucket{job=\"qwen-lab-engine\",model_name=\"qwen38-aibrix-lab\"}[1m])))",
          "legendFormat": "{{node_alias}}",
          "datasource": {
            "type": "prometheus",
            "uid": "prometheus"
          }
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "s",
          "custom": {
            "drawStyle": "line",
            "lineWidth": 2,
            "fillOpacity": 6,
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "table",
          "placement": "bottom",
          "calcs": [
            "lastNotNull",
            "max"
          ]
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 3,
      "title": "引擎请求平均 TPOT P95",
      "type": "timeseries",
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "gridPos": {
        "x": 12,
        "y": 0,
        "w": 12,
        "h": 9
      },
      "targets": [
        {
          "refId": "A",
          "expr": "histogram_quantile(0.95,sum by(le,node_alias)(rate(vllm:request_time_per_output_token_seconds_bucket{job=\"qwen-lab-engine\",model_name=\"qwen38-aibrix-lab\"}[1m])))",
          "legendFormat": "{{node_alias}}",
          "datasource": {
            "type": "prometheus",
            "uid": "prometheus"
          }
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "s",
          "custom": {
            "drawStyle": "line",
            "lineWidth": 2,
            "fillOpacity": 6,
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "table",
          "placement": "bottom",
          "calcs": [
            "lastNotNull",
            "max"
          ]
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 6,
      "title": "输出吞吐",
      "type": "timeseries",
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "gridPos": {
        "x": 0,
        "y": 9,
        "w": 12,
        "h": 9
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum by(node_alias)(rate(vllm:generation_tokens_total{job=\"qwen-lab-engine\",model_name=\"qwen38-aibrix-lab\"}[1m]))",
          "legendFormat": "{{node_alias}}",
          "datasource": {
            "type": "prometheus",
            "uid": "prometheus"
          }
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "suffix:Token/s",
          "custom": {
            "drawStyle": "line",
            "lineWidth": 2,
            "fillOpacity": 6,
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "table",
          "placement": "bottom",
          "calcs": [
            "lastNotNull",
            "max"
          ]
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 8,
      "title": "前缀 Token 命中率",
      "type": "timeseries",
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "gridPos": {
        "x": 12,
        "y": 9,
        "w": 12,
        "h": 9
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum by(node_alias)(rate(vllm:prefix_cache_hits_total{job=\"qwen-lab-engine\",model_name=\"qwen38-aibrix-lab\"}[1m])) / sum by(node_alias)(rate(vllm:prefix_cache_queries_total{job=\"qwen-lab-engine\",model_name=\"qwen38-aibrix-lab\"}[1m]))",
          "legendFormat": "{{node_alias}}",
          "datasource": {
            "type": "prometheus",
            "uid": "prometheus"
          }
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "percentunit",
          "custom": {
            "drawStyle": "line",
            "lineWidth": 2,
            "fillOpacity": 6,
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "table",
          "placement": "bottom",
          "calcs": [
            "lastNotNull",
            "max"
          ]
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    }
  ],
  "links": [],
  "templating": {
    "list": []
  },
  "annotations": {
    "list": [
      {
        "builtIn": 1,
        "datasource": {
          "type": "grafana",
          "uid": "-- Grafana --"
        },
        "enable": true,
        "hide": true,
        "iconColor": "#2563eb",
        "name": "Annotations & Alerts",
        "type": "dashboard"
      }
    ]
  }
}