我有一个具有以下结构的集合：
SELECT * INTO VendorsBackup
from Vendors
作为输入，我有一个{
"_id" : ..,
"conversationID" : "conversation_1_id",
"sender" : "user_2_Id",
"msgbody" : "Hi there !",
"ts" : "2019-03-25T23:12:14.235Z",
}
{
"_id" : ..,
"conversationID" : "conversation_1_id",
"sender" : "user_1_Id",
"msgbody" : "Hello !",
"ts" : "2019-03-26T02:45:44.627Z",
}
{
"_id" : ..,
"conversationID" : "conversation_2_id",
"sender" : "user_3_Id",
"msgbody" : "Hi !",
"ts" : "2019-03-26T02:45:44.627Z",
}
{
"_id" : ..,
"conversationID" : "conversation_2_id",
"sender" : "user_4_Id",
"msgbody" : "Good morning !",
"ts" : "2019-03-26T02:45:44.627Z",
}
数组，我需要到达每个请求的sessionID的最后一条消息（按ts字段排序）。
我应该使用$first (aggregation)方法还是这种查询的低性能解决方案？
有人可以帮我写一个这样的查询的例子吗？