Elasticsearch 中 Pipeline 聚合有哪些？

推荐答案

在 Elasticsearch 中，Pipeline 聚合是一种特殊的聚合类型，它允许你在其他聚合的结果上进行进一步的计算和处理。常见的 Pipeline 聚合包括：

1. Bucket Script Aggregation：允许你在多个桶聚合的结果上执行脚本计算。
2. Bucket Selector Aggregation：用于根据条件筛选桶。
3. Cumulative Sum Aggregation：计算累积和。
4. Derivative Aggregation：计算指标的导数。
5. Moving Function Aggregation：对时间序列数据应用移动窗口函数。
6. Normalize Aggregation：对指标进行归一化处理。
7. Serial Differencing Aggregation：计算时间序列数据的差分。
8. Stats Bucket Aggregation：计算多个桶的统计信息。
9. Sum Bucket Aggregation：对多个桶的值进行求和。
10. Extended Stats Bucket Aggregation：计算多个桶的扩展统计信息。
11. Percentiles Bucket Aggregation：计算多个桶的百分位数。
12. Bucket Sort Aggregation：对桶进行排序。

本题详细解读

1. Bucket Script Aggregation

Bucket Script Aggregation 允许你在多个桶聚合的结果上执行脚本计算。例如，你可以计算两个不同桶的值的差值。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_diff": {
          "bucket_script": {
            "buckets_path": {
              "prev_sales": "total_sales"
            },
            "script": "params.prev_sales"
          }
        }
      }
    }
  }
}

2. Bucket Selector Aggregation

Bucket Selector Aggregation 用于根据条件筛选桶。例如，你可以筛选出销售额大于某个值的桶。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_bucket_filter": {
          "bucket_selector": {
            "buckets_path": {
              "totalSales": "total_sales"
            },
            "script": "params.totalSales > 1000"
          }
        }
      }
    }
  }
}

3. Cumulative Sum Aggregation

Cumulative Sum Aggregation 用于计算累积和。例如，你可以计算每个月的销售额的累积和。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "cumulative_sales": {
          "cumulative_sum": {
            "buckets_path": "total_sales"
          }
        }
      }
    }
  }
}

4. Derivative Aggregation

Derivative Aggregation 用于计算指标的导数。例如，你可以计算销售额的变化率。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_derivative": {
          "derivative": {
            "buckets_path": "total_sales"
          }
        }
      }
    }
  }
}

5. Moving Function Aggregation

Moving Function Aggregation 用于对时间序列数据应用移动窗口函数。例如，你可以计算销售额的移动平均值。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "moving_avg": {
          "moving_fn": {
            "buckets_path": "total_sales",
            "window": 3,
            "script": "MovingFunctions.unweightedAvg(values)"
          }
        }
      }
    }
  }
}

6. Normalize Aggregation

Normalize Aggregation 用于对指标进行归一化处理。例如，你可以将销售额归一化到 0 到 1 之间。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "normalized_sales": {
          "normalize": {
            "buckets_path": "total_sales",
            "method": "rescale_0_1"
          }
        }
      }
    }
  }
}

7. Serial Differencing Aggregation

Serial Differencing Aggregation 用于计算时间序列数据的差分。例如，你可以计算销售额的差分。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_diff": {
          "serial_diff": {
            "buckets_path": "total_sales",
            "lag": 1
          }
        }
      }
    }
  }
}

8. Stats Bucket Aggregation

Stats Bucket Aggregation 用于计算多个桶的统计信息。例如，你可以计算每个月的销售额的统计信息。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_stats": {
          "stats_bucket": {
            "buckets_path": "total_sales"
          }
        }
      }
    }
  }
}

9. Sum Bucket Aggregation

Sum Bucket Aggregation 用于对多个桶的值进行求和。例如，你可以计算所有月份的销售额总和。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        }
      }
    },
    "total_sales_sum": {
      "sum_bucket": {
        "buckets_path": "sales_per_month>total_sales"
      }
    }
  }
}

10. Extended Stats Bucket Aggregation

Extended Stats Bucket Aggregation 用于计算多个桶的扩展统计信息。例如，你可以计算每个月的销售额的扩展统计信息。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_extended_stats": {
          "extended_stats_bucket": {
            "buckets_path": "total_sales"
          }
        }
      }
    }
  }
}

11. Percentiles Bucket Aggregation

Percentiles Bucket Aggregation 用于计算多个桶的百分位数。例如，你可以计算每个月的销售额的百分位数。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_percentiles": {
          "percentiles_bucket": {
            "buckets_path": "total_sales",
            "percents": [25, 50, 75]
          }
        }
      }
    }
  }
}

12. Bucket Sort Aggregation

Bucket Sort Aggregation 用于对桶进行排序。例如，你可以按销售额对月份进行排序。

// www.javascriptcn.com code example
{
  "aggs": {
    "sales_per_month": {
      "date_histogram": {
        "field": "date",
        "calendar_interval": "month"
      },
      "aggs": {
        "total_sales": {
          "sum": {
            "field": "price"
          }
        },
        "sales_bucket_sort": {
          "bucket_sort": {
            "sort": [
              {
                "total_sales": {
                  "order": "desc"
                }
              }
            ],
            "size": 10
          }
        }
      }
    }
  }
}