嵌套聚合与子聚合
学习通过将指标嵌套在分桶中组合聚合，构建多层次分析，并使用嵌套聚合遍历嵌套的文档结构。
嵌套聚合与子聚合 是 CoddyKit 上的免费 Elasticsearch & Full Text Search Systems 课时。 这是第 4 节课，共 4 节。 你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。 这是 Elasticsearch & Full Text Search Systems 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 Elasticsearch & Full Text Search Systems 课程共包含 4 节课。
本课时的部分内容尚未翻译，以英文显示。
Combining Aggregations
A single aggregation answers one question. Real analytics often need layered answers: average price per category, or top brands per region. Elasticsearch lets you nest aggregations inside one another.
This lesson shows how to compose them.
The aggs Hierarchy
Any bucket aggregation can contain a sub-aggs block. The sub-aggregation runs once per bucket, operating only on the documents in that bucket.
This is the core mechanism for multi-dimensional analytics.
Metric Inside a Bucket
Here we group documents by category, then compute the average price within each group. The metric aggregation lives inside the bucket aggregation's aggs.
GET sales/_search
{
"size": 0,
"aggs": {
"by_category": {
"terms": { "field": "category" },
"aggs": {
"avg_price": { "avg": { "field": "price" } }
}
}
}
}Reading the Result
Each bucket in by_category now carries an avg_price value alongside its doc_count. You get the category name, how many docs it holds, and its average price in one response.
Bucket Inside a Bucket
You can nest bucket aggregations too. Group by region, then by brand within each region. This produces a two-level breakdown.
"aggs": {
"by_region": {
"terms": { "field": "region" },
"aggs": {
"by_brand": { "terms": { "field": "brand" } }
}
}
}Mind the Cardinality
Deeply nested terms aggregations can explode combinatorially. A region with 50 brands across 20 regions yields up to 1,000 buckets. Use the size parameter to limit returned buckets and protect memory.
Multiple Sub-Aggregations
A bucket can hold several sub-aggregations at once. Here each category reports both its average and maximum price.
"by_category": {
"terms": { "field": "category" },
"aggs": {
"avg_price": { "avg": { "field": "price" } },
"max_price": { "max": { "field": "price" } }
}
}The nested Aggregation
When data uses the nested field type, you must enter that scope with a nested aggregation before aggregating on its inner fields. It points at a path.
"aggs": {
"variants": {
"nested": { "path": "variants" },
"aggs": {
"avg_qty": { "avg": { "field": "variants.quantity" } }
}
}
}Reverse Nested
Inside a nested aggregation you can jump back to the parent document scope using reverse_nested. This lets you count distinct parent docs that contain a matching nested element.
"reverse_nested": {},
"aggs": {
"parent_count": { "value_count": { "field": "_id" } }
}Sorting Buckets by a Metric
Order parent buckets by a nested sub-metric using order. Here categories are sorted by their average price, descending.
"terms": {
"field": "category",
"order": { "avg_price": "desc" }
}Design Tips
Keep nesting shallow when possible, always set size on inner terms, and put the most selective bucket aggregation first to reduce the work done by deeper levels.
Quick Check
Test your grasp of sub-aggregations.
Recap
You learned to compose aggregations:
- Any bucket aggregation can hold sub-
aggsthat run per bucket. - Nest metrics in buckets, or buckets in buckets for multi-level breakdowns.
- Use the
nestedandreverse_nestedaggregations to traverse nested document scopes. - Control bucket explosion with
sizeand order buckets by sub-metrics withorder.
用 AI 导师学习 Elasticsearch & Full Text Search Systems — 免费
在浏览器中编写并运行真实代码，获得全天候 AI 导师的即时帮助，并在网页或应用中继续学习。
- 课程
- 12
- 课程
- 48
常见问题解答
「嵌套聚合与子聚合」课时是免费的吗？
是的 — 「嵌套聚合与子聚合」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 Elasticsearch & Full Text Search Systems 课程的其余内容，请升级到 CoddyKit PRO。 Elasticsearch & Full Text Search Systems 课程共包含 4 节课。
「嵌套聚合与子聚合」这节课中我会学到什么？
学习通过将指标嵌套在分桶中组合聚合，构建多层次分析，并使用嵌套聚合遍历嵌套的文档结构。 你通过在浏览器中直接运行的动手代码来练习 Elasticsearch & Full Text Search Systems，全天候 AI 导师会在你学习这节课的过程中回答你的问题。
学习 Elasticsearch & Full Text Search Systems 需要有经验吗？
无需任何先前经验。CoddyKit 上的 Elasticsearch & Full Text Search Systems 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。 这是第 4 节课，共 4 节。
「嵌套聚合与子聚合」课时需要多长时间？
大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。
我能在这节 Elasticsearch & Full Text Search Systems 课中编写并运行代码吗？
能。每节 Elasticsearch & Full Text Search Systems 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。