Elasticsearch：FalconSeek向量索引使用指南

本文檔詳細介紹FalconSeek在向量檢索情境中的進階用法，協助您根據需求（如大規模資料存放區成本敏感、極致效能最佳化等）選擇HNSW、RabitQGraph等演算法進行加速查詢。涵蓋演算法選型指南、參數配置、索引管理及動態調優技巧，並提供完整程式碼範例，協助實現高效向量檢索。

背景資訊

FalconSeek在阿里雲ES原有索引結構之上，增加了基於C++的向量引擎全新索引，FalconSeek向量索引是阿里巴巴自主研發，支援了阿里巴巴集團內淘寶、天貓搜尋、推薦、拍立淘等集團內的主要業務，您可以利用 FalconSeek 核心整合的高效能向量索引，構建高效的以圖搜圖、語義搜尋等 AI 應用。

FalconSeek 向量索引完全相容開源 ES 向量引擎功能（如kNN 搜尋等）。在阿里雲ES索引中使用FalconSeek僅需一步：在索引配置中將 index_options.type 設定為 havenask_native，更多進階參數配置請參見下文詳細說明。

使用樣本

基礎樣本

以下樣本建立一個儲存和檢索向量資料的索引，索引名為 my_falcon_seek_index ，其中包含一個名為 product_vector 的向量欄位，維度為 128，並使用HNSW 演算法。

1. 建立索引

   PUT /my_falcon_seek_index
   {
     "settings": {
       "number_of_shards": 1,
       "number_of_replicas": 0
     },
     "mappings": {
       "properties": {
         "product_vector": {
           "type": "dense_vector",
           "dims": 128,
           "index": true,
           "similarity": "l2_norm",
           "index_options": {
             "type": "havenask_native",
             "knn_type": "HNSW",
             "m": 32,
             "ef_construction": 400
           }
         },
         "category": {
           "type": "keyword" 
         }
       }
     }
   }

   核心參數：

   • dense_vector 欄位類型，是用於儲存稠密向量資料的專用欄位類型。在 mappings 中定義此類型欄位時，必須指定以下核心屬性：

     • dims：向量的維度。

     • similarity：向量間的相似性計算函數。

     • index_options：向量索引的詳細配置，包括演算法類型和相關參數。

   • similarity 相似性函數，用于衡量兩個向量之間的相似程度，選擇合適的函數對召回效果至關重要。

     函數	說明	適用情境
     l2_norm	歐氏距離。計算兩個向量在多維空間中的直線距離。距離越小，越相似。	通用情境，如Image Recognition、Face Service。
     cosine	餘弦相似性。計算兩個向量方向的夾角餘弦值。值越接近 1，越相似。	文本語義相似性分析，不受向量長度影響。
     dot_product	內積。計算兩個向量的點積。值越大，越相似。	推薦系統等需要考慮向量模長的情境。
     max_inner_product	與dot_product功能一樣，但不要求向量歸一化。	推薦系統等需要考慮向量模長的情境。

2. 寫入資料：向索引中寫入包含向量和中繼資料（如商品類別）的文檔，product_vector 數組的長度必須與 dims 定義的維度（128）完全一致。

   POST /my_falcon_seek_index/_doc/1
   {
     "product_vector": [0.12, -0.05, 0.08, 0.24, -0.17, 0.31, 0.02, -0.19, 0.11, 0.28,
       -0.03, 0.15, 0.22, -0.11, 0.09, 0.33, -0.07, 0.14, 0.26, -0.21,
       0.18, 0.29, -0.13, 0.06, 0.35, -0.08, 0.16, 0.23, -0.15, 0.12, 
       0.27, -0.22, 0.19, 0.32, -0.14, 0.07, 0.25, -0.18, 0.13, 0.30,
       -0.09, 0.17, 0.24, -0.16, 0.10, 0.34, -0.10, 0.20, 0.31, -0.23,
       0.15, 0.28, -0.12, 0.11, 0.26, -0.19, 0.14, 0.29, -0.17, 0.08,
       0.22, -0.20, 0.16, 0.27, -0.15, 0.09, 0.25, -0.21, 0.18, 0.30,
       -0.13, 0.07, 0.24, -0.22, 0.19, 0.32, -0.16, 0.10, 0.26, -0.18,
       0.12, 0.28, -0.14, 0.06, 0.23, -0.19, 0.15, 0.29, -0.11, 0.05,
       0.21, -0.17, 0.13, 0.27, -0.10, 0.04, 0.20, -0.15, 0.11, 0.25,
       -0.09, 0.03, 0.19, -0.13, 0.10, 0.24, -0.08, 0.02, 0.18, -0.12,
       0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.22, 0.05, -0.06,
       0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.22],
     "category": "clothes"
   }
   POST /my_falcon_seek_index/_doc/2
   {"product_vector":[0.12, -0.05, 0.08, 0.24, -0.17, 0.31, 0.02, -0.19, 0.11, 0.28,
       -0.03, 0.15, 0.22, -0.11, 0.09, 0.33, -0.07, 0.14, 0.26, -0.21,
       0.18, 0.29, -0.13, 0.06, 0.35, -0.08, 0.16, 0.23, -0.15, 0.12, 
       0.27, -0.22, 0.19, 0.32, -0.14, 0.07, 0.25, -0.18, 0.13, 0.30,
       -0.09, 0.17, 0.24, -0.16, 0.10, 0.34, -0.10, 0.20, 0.31, -0.23,
       0.15, 0.28, -0.12, 0.11, 0.26, -0.19, 0.14, 0.29, -0.17, 0.08,
       0.22, -0.20, 0.16, 0.27, -0.15, 0.09, 0.25, -0.21, 0.18, 0.30,
       -0.13, 0.07, 0.24, -0.22, 0.19, 0.32, -0.16, 0.10, 0.26, -0.18,
       0.12, 0.28, -0.14, 0.06, 0.23, -0.19, 0.15, 0.29, -0.11, 0.05,
       0.21, -0.17, 0.13, 0.27, -0.10, 0.04, 0.20, -0.15, 0.11, 0.25,
       -0.09, 0.03, 0.19, -0.13, 0.10, 0.24, -0.08, 0.02, 0.18, -0.12,
       0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.22, 0.05, -0.06,
       0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.22],"category": "clothes"}

3. 向量檢索（k-NN）： 根據給定的查詢向量，尋找最相似的 5個文檔。

   GET /my_falcon_seek_index/_search
   {
     "knn": {
       "field": "product_vector",
       "query_vector": [0.12, -0.05, 0.01, 0.24, -0.17, 0.31, 0.02, -0.19, 0.11, 0.28,
       -0.03, 0.15, 0.22, -0.11, 0.09, 0.23, -0.07, 0.14, 0.26, -0.21,
       0.18, 0.29, -0.13, 0.06, 0.35, -0.18, 0.16, 0.23, -0.15, 0.12, 
       0.27, -0.22, 0.19, 0.32, -0.14, 0.87, 0.25, -0.18, 0.13, 0.30,
       -0.09, 0.17, 0.24, -0.16, 0.10, 0.64, -0.10, 0.20, 0.31, -0.23,
       0.15, 0.28, -0.12, 0.11, 0.26, -0.19, 0.14, 0.29, -0.17, 0.08,
       0.22, -0.20, 0.16, 0.27, -0.15, 0.09, 0.25, -0.21, 0.18, 0.30,
       -0.13, 0.07, 0.24, -0.22, 0.19, 0.52, -0.16, 0.10, 0.26, -0.18,
       0.12, 0.28, -0.14, 0.06, 0.23, -0.19, 0.14, 0.29, -0.11, 0.05,
       0.21, -0.17, 0.13, 0.27, -0.10, 0.04, 0.20, -0.15, 0.11, 0.25,
       -0.09, 0.03, 0.19, -0.13, 0.10, 0.24, -0.08, 0.02, 0.18, -0.12,
       0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.22, 0.05, -0.06,
       0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.12],
       "k": 5,
       "num_candidates": 100
     }
   }

   核心參數：

   knn 查詢文法，是執行 k-最近鄰（k-Nearest Neighbors）搜尋的專用查詢體。

   • field：要查詢的 dense_vector 欄位名。

   • query_vector：用於查詢的向量，其維度必須與欄位定義一致。

   • k：需要返回的最相似結果數量。

   • num_candidates：在每個分區上，演算法內部搜尋的候選集大小。該值大於 k，通常是 k 的數倍。值越大，召回率越高，但查詢延遲也相應增加。

4. 過濾檢索：在執行向量檢索的同時，對結果進行過濾，以滿足更複雜的業務需求。以下樣本將在 category 為 "shoes" 的文檔中，尋找最相似的 5 個文檔。

   GET /my_falcon_seek_index/_search
   {
     "knn": {
       "field": "product_vector",
       "query_vector": [
         0.12, -0.05, 0.01, 0.24, -0.17, 0.31, 0.02, -0.19, 0.11, 0.28,
         -0.03, 0.15, 0.22, -0.11, 0.09, 0.23, -0.07, 0.14, 0.26, -0.21,
         0.18, 0.29, -0.13, 0.06, 0.35, -0.18, 0.16, 0.23, -0.15, 0.12,
         0.27, -0.22, 0.19, 0.32, -0.14, 0.87, 0.25, -0.18, 0.13, 0.30,
         -0.09, 0.17, 0.24, -0.16, 0.10, 0.64, -0.10, 0.20, 0.31, -0.23,
         0.15, 0.28, -0.12, 0.11, 0.26, -0.19, 0.14, 0.29, -0.17, 0.08,
         0.22, -0.20, 0.16, 0.27, -0.15, 0.09, 0.25, -0.21, 0.18, 0.30,
         -0.13, 0.07, 0.24, -0.22, 0.19, 0.52, -0.16, 0.10, 0.26, -0.18,
         0.12, 0.28, -0.14, 0.06, 0.23, -0.19, 0.14, 0.29, -0.11, 0.05,
         0.21, -0.17, 0.13, 0.27, -0.10, 0.04, 0.20, -0.15, 0.11, 0.25,
         -0.09, 0.03, 0.19, -0.13, 0.10, 0.24, -0.08, 0.02, 0.18, -0.12,
         0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.22, 0.05, -0.06,
         0.09, 0.23, -0.07, 0.01, 0.17, -0.11, 0.08, 0.12
       ],
       "k": 5,
       "num_candidates": 100,
       "filter": {
         "term": {
           "category": "shoes"
         }
       }
     }
   }

擴充功能：tags_filter過濾檢索

tags_filter 是一種專為 HNSW 和 QGraph 演算法最佳化的前置過濾機制。與標準的 bool filter 相比，它在圖遍曆的早期階段就排除不匹配的節點，因此效能更高。

• 適用情境：當過濾欄位取值有限（如類目、品牌 ID），且對查詢效能要求極致時。

• 如何啟用：

  1. 在 mappings 的 index_options 中，使用 tags 參數聲明要用於過濾的 keyword 類型欄位。

     PUT /my_vector_index_with_tags
     {
       "mappings": {
         "properties": {
           "product_vector": {
             "type": "dense_vector",
             "dims": 128,
             "index_options": {
               "type": "havenask_native",
               "knn_type": "HNSW",
               "tags": ["category"] // 聲明 category 欄位用於 tags_filter
             }
           },
           "category": {
             "type": "keyword" // 必須是 keyword 類型
           }
         }
       }
     }

  2. 在查詢的 knn 體中使用 tags_filter 參數，並採用 "field_name = value" 文法。支援 | (OR) 和 & (AND) 邏輯。

     GET /my_vector_index_with_tags/_search
     {
       "knn": {
         "field": "product_vector",
         "query_vector": [...],
         "k": 5,
         "tags_filter": "category = shoes | category = socks"
       }
     }
     

在建立索引時，所有向量相關的配置都在 mappings 的 index_options 塊內完成。

PUT /<your_index_name>
{
  "mappings": {
    "properties": {
      "<your_vector_field>": {
        "type": "dense_vector",
        "dims": 768,
        "similarity": "l2_norm",
        "index_options": {
          // --- 通用參數 ---
          "type": "havenask_native",
          "knn_type": "HNSW",
          // --- 演算法構建參數 ---
          "m": 32,
          "ef_construction": 400,
          // --- 進階參數 ---
          "thread_count": 8
        }
      }
    }
  }
}

json

"index_options": {  "knn_type": "HNSW",  "m": 32, // 會被下面的 index_params 覆蓋  "index_params": {    "proxima.hnsw.builder.max_neighbor_count": 48 // 最終生效的是 48  }}

// 基礎 HNSW 配置
PUT /hnsw_basic
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 768,
        "index": true,
        "similarity": "cosine",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "HNSW"
        }
      }
    }
  }
}

// 高效能 HNSW 配置
PUT /hnsw_performance
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 1024,
        "index": true,
        "similarity": "dot_product",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "HNSW",
          "m": 48,
          "ef_construction": 500,
          "thread_count": 8,
          "linear_build_threshold": 1000,
          "is_embedding_saved": true,
          "embedding_load_strategy": "ANN_INDEX_FILE",
          "index_load_strategy": "MEM"
        }
      }
    }
  }
}

// 高效能配置
{
  "index_params": {
    "param.rabitQGraph.builder.neighbor_cnt": 256,
    "param.rabitQGraph.builder.ef_construction": 512,
    "param.rabitQGraph.builder.quantized_bit_count": 8,
    "param.rabitQGraph.builder.cluster_count": 128,
    "param.rabitQGraph.builder.thread_count": 8,
    "param.rabitQGraph.searcher.ef": 300,
    "param.rabitQGraph.searcher.max_scan_ratio": 0.1
  }
}

// 記憶體最佳化配置
{
  "index_params": {
    "param.rabitQGraph.builder.neighbor_cnt": 64,
    "param.rabitQGraph.builder.ef_construction": 200,
    "param.rabitQGraph.builder.quantized_bit_count": 1,
    "param.rabitQGraph.builder.cluster_count": 32,
    "param.rabitQGraph.searcher.ef": 150,
    "param.rabitQGraph.searcher.max_scan_ratio": 0.03
  }
}

// 平衡配置
{
  "index_params": {
    "param.rabitQGraph.builder.neighbor_cnt": 128,
    "param.rabitQGraph.builder.ef_construction": 400,
    "param.rabitQGraph.builder.quantized_bit_count": 4,
    "param.rabitQGraph.builder.cluster_count": 64,
    "param.rabitQGraph.searcher.ef": 250
  }
}

// 基礎配置
{
  "index_params": {
    "proxima.linear.builder.column_major_order": "false",
    "proxima.linear.searcher.read_block_size": 1048576
  }
}

// 大記憶體配置
{
  "index_params": {
    "proxima.linear.builder.column_major_order": "true",
    "proxima.linear.searcher.read_block_size": 2097152
  }
}

// 基礎配置
{
  "index_params": {
    "proxima.qc.builder.thread_count": 4,
    "proxima.qc.builder.centroid_count": "1000",
    "proxima.qc.builder.cluster_class": "OptKmeansCluster",
    "proxima.qc.searcher.scan_ratio": 0.02
  }
}

// 層次聚類配置
{
  "index_params": {
    "proxima.qc.builder.thread_count": 8,
    "proxima.qc.builder.centroid_count": "100*100",
    "proxima.qc.builder.optimizer_class": "HnswBuilder",
    "proxima.qc.builder.quantizer_class": "Int8QuantizerConverter",
    "proxima.qc.searcher.scan_ratio": 0.01
  }
}

// 高精度配置
{
  "index_params": {
    "proxima.qc.builder.thread_count": 12,
    "proxima.qc.builder.centroid_count": "2000",
    "proxima.qc.builder.train_sample_count": 100000,
    "proxima.qc.builder.store_original_features": true,
    "proxima.qc.searcher.scan_ratio": 0.05
  }
}

// Int8 量化配置
{
  "index_params": {
    "proxima.hnsw.builder.max_neighbor_count": 32,
    "proxima.hnsw.builder.efconstruction": 400,
    "proxima.hnsw.builder.thread_count": 4,
    "proxima.qgraph.builder.quantizer_class": "Int8QuantizerConverter",
    "proxima.qgraph.builder.quantizer_params": {},
    "proxima.hnsw.searcher.ef": 300
  }
}

// Int4 量化配置（更省記憶體）
{
  "index_params": {
    "proxima.hnsw.builder.max_neighbor_count": 48,
    "proxima.hnsw.builder.efconstruction": 500,
    "proxima.hnsw.builder.thread_count": 6,
    "proxima.qgraph.builder.quantizer_class": "Int4QuantizerConverter",
    "proxima.qgraph.builder.quantizer_params": {},
    "proxima.hnsw.searcher.ef": 400,
    "proxima.hnsw.searcher.max_scan_ratio": 0.1
  }
}

// HalfFloat 量化配置（高精度）
{
  "index_params": {
    "proxima.hnsw.builder.max_neighbor_count": 64,
    "proxima.hnsw.builder.efconstruction": 600,
    "proxima.hnsw.builder.thread_count": 8,
    "proxima.qgraph.builder.quantizer_class": "HalfFloatConverter",
    "proxima.qgraph.builder.quantizer_params": {},
    "proxima.hnsw.searcher.ef": 500
  }
}

{
  "search_params": {
    "proxima.hnsw.searcher.ef": "400",
    "proxima.hnsw.searcher.max_scan_ratio": "0.15"
  }
}

{
  "search_params": {
    "proxima.hnsw.searcher.ef": "400",
    "proxima.hnsw.searcher.max_scan_ratio": "0.15"
  }
}

{
  "search_params": {
    "proxima.qc.searcher.scan_ratio": "0.02"
  }
}

{
  "search_params": {
    "param.rabitQGraph.searcher.ef": "300",
    "param.rabitQGraph.searcher.max_scan_ratio": "0.08"
  }
}

// HNSW 動態調整精度和效能
GET vector_index/_search
{
  "knn": {
    "field": "vector",
    "query_vector": [0.1, 0.2, 0.3],
    "k": 10,
    "num_candidates": 100,
    "search_params": {
      "proxima.hnsw.searcher.ef": "500",
      "proxima.hnsw.searcher.max_scan_ratio": "0.2"
    }
  }
}

// QC 動態調整掃描比例
GET vector_index/_search
{
  "knn": {
    "field": "vector",
    "query_vector": [0.1, 0.2, 0.3],
    "k": 10,
    "num_candidates": 100,
    "search_params": {
      "proxima.qc.searcher.scan_ratio": "0.05"
    }
  }
}

// RabitQGraph 動態調整搜尋精度
GET vector_index/_search
{
  "knn": {
    "field": "vector",
    "query_vector": [0.1, 0.2, 0.3],
    "k": 10,
    "num_candidates": 100,
    "search_params": {
      "param.rabitQGraph.searcher.ef": "400",
      "param.rabitQGraph.searcher.max_scan_ratio": "0.1"
    }
  }
}

// 禁用線性閾值
{
  "linear_build_threshold": 0
}

// 小資料集使用線性搜尋
{
  "linear_build_threshold": 1000
}

// 較大閾值，適用於測試環境
{
  "linear_build_threshold": 5000
}

附錄：完整樣本

// 基礎 HNSW 配置
PUT /hnsw_basic
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 768,
        "index": true,
        "similarity": "cosine",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "HNSW"
        }
      }
    }
  }
}

// 高效能 HNSW 配置
PUT /hnsw_performance
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 1024,
        "index": true,
        "similarity": "dot_product",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "HNSW",
          "m": 48,
          "ef_construction": 500,
          "thread_count": 8,
          "linear_build_threshold": 1000,
          "is_embedding_saved": true,
          "embedding_load_strategy": "ANN_INDEX_FILE",
          "index_load_strategy": "MEM"
        }
      }
    }
  }
}

// 基礎 Linear 配置
PUT /linear_basic
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 384,
        "index": true,
        "similarity": "l2_norm",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "Linear"
        }
      }
    }
  }
}

// 基礎 QC 配置
PUT /qc_basic
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 768,
        "index": true,
        "similarity": "cosine",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "QC"
        }
      }
    }
  }
}

// 自訂 QC 配置
PUT /qc_custom
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 1024,
        "index": true,
        "similarity": "dot_product",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "QC",
          "thread_count": 8,
          "linear_build_threshold": 5000,
          "index_params": "{\"proxima.qc.builder.thread_count\": 8, \"proxima.qc.builder.centroid_count\": \"2000\"}"
        }
      }
    }
  }
}

// 基礎 QGraph 配置
PUT /qgraph_basic
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 768,
        "index": true,
        "similarity": "cosine",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "QGraph",
          "quantizer": "int8"
        }
      }
    }
  }
}

// 高精度 QGraph 配置
PUT /qgraph_high_precision
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 1536,
        "index": true,
        "similarity": "cosine",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "QGraph",
          "m": 40,
          "ef_construction": 600,
          "thread_count": 8,
          "quantizer": "fp16",
          "is_embedding_saved": true,
          "index_load_strategy": "MEM"
        }
      }
    }
  }
}

// 記憶體最佳化 QGraph 配置
PUT /qgraph_memory_optimized
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 1024,
        "index": true,
        "similarity": "l2_norm",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "QGraph",
          "m": 24,
          "ef_construction": 300,
          "thread_count": 6,
          "quantizer": "int4",
          "is_embedding_saved": false,
          "index_load_strategy": "BUFFER"
        }
      }
    }
  }
}

// 基礎 RabitQGraph 配置（僅支援 l2_norm 相似性）
PUT /rabitqgraph_basic
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 64,
        "index": true,
        "similarity": "l2_norm",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "RabitQGraph"
        }
      }
    }
  }
}

// 高效能 RabitQGraph 配置 - 使用 Map 格式 index_params
PUT /rabitqgraph_performance
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 128,
        "index": true,
        "similarity": "l2_norm",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "RabitQGraph",
          "thread_count": 8,
          "index_params": {
            "param.rabitQGraph.builder.neighbor_cnt": 256,
            "param.rabitQGraph.builder.ef_construction": 512,
            "param.rabitQGraph.builder.quantized_bit_count": 4,
            "param.rabitQGraph.builder.cluster_count": 128,
            "param.rabitQGraph.searcher.ef": 300
          }
        }
      }
    }
  }
}

// 記憶體最佳化 RabitQGraph 配置
PUT /rabitqgraph_memory_optimized
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 192,
        "index": true,
        "similarity": "l2_norm",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "RabitQGraph",
          "thread_count": 4,
          "linear_build_threshold": 1000,
          "index_params": {
            "param.rabitQGraph.builder.neighbor_cnt": 64,
            "param.rabitQGraph.builder.ef_construction": 200,
            "param.rabitQGraph.builder.quantized_bit_count": 1,
            "param.rabitQGraph.builder.cluster_count": 32,
            "param.rabitQGraph.searcher.ef": 150,
            "param.rabitQGraph.searcher.max_scan_ratio": 0.05
          }
        }
      }
    }
  }
}

// RabitQGraph 帶標籤過濾配置
PUT /rabitqgraph_with_tags
{
  "mappings": {
    "properties": {
      "vector": {
        "type": "dense_vector",
        "dims": 256,
        "index": true,
        "similarity": "l2_norm",
        "index_options": {
          "type": "havenask_native",
          "knn_type": "RabitQGraph",
          "tags": ["category", "region"],
          "index_params": {
            "param.rabitQGraph.builder.neighbor_cnt": 128,
            "param.rabitQGraph.builder.ef_construction": 400,
            "param.rabitQGraph.builder.quantized_bit_count": 8,
            "param.rabitQGraph.searcher.ef": 250
          }
        }
      },
      "category": {
        "type": "keyword"
      },
      "region": {
        "type": "keyword"
      }
    }
  }
}