Question

I am currently evaluating the efficiency of different databases for a use case. In Mongodb, would like to store around 1 million objects with the following structure. Each object will have between 5 and 10 objects in the foo array.

{
   name:"my name",
   foos:[
      {
         foo:"...",
         bar:"..."
      },
      {
         foo:"...",
         bar:"..."
      },
      {
         foo:"...",
         bar:"..."
      }
   ]
}

I often need to search for objects which where the foos collection contains an object with a specific property, e.g.:

// mongo collection
[
   {
      name:'my name',
      foos:[
         {
            foo:'one_foo',
            bar:'a_bar'
         },
         {
            foo:'two_foo',
            bar:'b_bar'
         }
      ]
   },
   {
      name:'another name',
      foos:[
         {
            foo:'another foo',
            bar:'a_bar'
         },
         {
            foo:'just another foo',
            bar:'c_bar'
         }
      ]
   }
]

// search (pseudo code)
{ foos: {$elemMatch: {bar: 'c_bar'}} }
// returns
{
      name:'another name',
      foos:[
         {
            foo:'another foo',
            bar:'a_bar'
         },
         {
            foo:'just another foo',
            bar:'c_bar'
         }
      ]
   }

Can this efficiently be done with mongo and how should the indexes be set?
I don’t want you to evaluate performance for me, just an idea how mongo performs for my use case or how optimization could look like.

Answer 1

MongoDB has documentation explaining how to create indexes on embedded documents, through dot notation:

Dot Notation (Reaching into Objects)

> db.blogposts.findOne()
{ title : "My First Post", author: "Jane",
  comments : [{ by: "Abe", text: "First" },
              { by : "Ada", text : "Good post" } ]
}
> db.blogposts.find( { "comments.by" : "Ada" } )

> db.blogposts.ensureIndex( { "comments.by" : 1 } );

As for the performance characteristic… just test it with your dataset.