我有一个索引“测试”。文档结构如下图所示。每个文档都有一个“标签”数组。我不知道如何查询此索引来获取前 10 个最常出现的标签？
\n\n另外，如果该索引中有超过 200 万篇文档，那么应该遵循哪些最佳实践？
\n\n{\n "_index" : "test",\n "_type" : "data",\n "_id" : "1412879673545024927_1373991666",\n "_score" : 1.0,\n "_source" : {\n "instagramuserid" : "1373991666",\n "likes_count" : 163,\n "@timestamp" : "2017-06-08T08:52:41.803Z",\n "post" : {\n "created_time" : "1482648403",\n "comments" : {\n "count" : 9\n },\n "user_has_liked" : true,\n "link" : "https://www.instagram.com/p/BObjpPMBWWf/",\n "caption" : {\n "created_time" : "1482648403",\n "from" : {\n "full_name" : "PARAMSahib \xe2\x84\xa2",\n "profile_picture" : "https://scontent.cdninstagram.com/t51.2885-19/s150x150/12750236_1692144537739696_350427084_a.jpg",\n "id" : "1373991666",\n "username" : "parambanana"\n },\n "id" : "17845953787172829",\n "text" : …