I have two collections, they both have one common field:
collection1
{
"myCommonFieldId" : 1
}
{
"myCommonFieldId" : 2
}
{
"myCommonFieldId" : 3
}
{
"myCommonFieldId" : 4
}
{
"myCommonFieldId" : 5
}
collection2
{
"myCommonFieldId" : 2
}
{
"myCommonFieldId" : 5
}
I want to get documents from collection1 which are not in collections2 (The uncommon myCommonFieldId-s between two collections). In this case:
{
"myCommonFieldId" : 1
}
{
"myCommonFieldId" : 3
}
{
"myCommonFieldId" : 4
}
The straightforward solution is to query collection2, take the results and run $nin in collection1
db.collection2.find({});
db.collection1.find({"myCommonFieldId" : {$nin : [<Result from query1>]}})
I have a huge dataset and since $nin operator is not efficient, $nin is becoming very expensive operation for me: Quoting from MongoDB documentations:
The inequality operator $nin is not very selective since it often matches a large portion of the index. As a result, in many cases, a $nin query with an index may perform no better than a $nin query that must scan all documents in a collection.
Is there any efficient way to do this?