聚合主要有三大類： 

A family of aggregations that build buckets, where each bucket is associated with a key and a document criterion. When the aggregation is executed, all the buckets criteria are evaluated on every document in the context and when a criterion matches, the document is considered to "fall in" the relevant bucket. By the end of the aggregation process, we’ll end up with a list of buckets - each one with a set of documents that "belong" to it. 比如：屬性的值得範圍，分類之類的 Aggregations that keep track and compute metrics over a set of documents. 和數學運算相關，比如：求取平均值 Aggregations that aggregate the output of other aggregations and their associated metrics 關聯其他聚合

SearchResponse sr = node.client().prepareSearch()
    .addAggregation(
        AggregationBuilders.terms("by_country").field("country")
        .subAggregation(AggregationBuilders.dateHistogram("by_year")
            .field("dateOfBirth")
            .interval((DateHistogramInterval.YEAR)
            .subAggregation(AggregationBuilders.avg("avg_children").field("children"))
        )
    )
    .execute().actionGet();
 

• Terms aggregation (bucket)
  Date Histogram aggregation (bucket)
  Average aggregation (metric)

     

 metric Aggregation幾具體操作：

max、min、sun、avg

MetricsAggregationBuilder aggregation =
        AggregationBuilders
                .avg("agg")
                .field("height");

取結果

Avg agg = sr.getAggregations().get("agg");
double value = agg.getValue();
 

stats包括上述的外加一個count

而extende stats 不僅包括stats還包括平法差一類的，更全面

還有百分之幾，範圍什麼的，可以根據自己的需求再看看這一方面的介紹，metric Aggregation

Bucket aggregations

global 全域性的，filter 、filters、missing 沒有值或者是null的統計、terms 分類、range、date range、histogram 直方圖、date histogram 等等，具體的運用看Bucket aggregation的介紹。