In this paper we will examine the key features of the database management system MongoDB. We will focus on the basic operations of CRUD and indexes. For our example we will create two databases one using MySQL and one in MongoDB. We will also compare the way that data will be created, selected, inserted and deleted in both databases. For the index part we will talk about the different types used in MongoDB comparing them with the indexes used in a relational database. Index Terms - MongoDB, NoSQL, BSON, CRUD, index. I. Introduction MongoDB is a high performance and very scalable document-oriented database developed in C++ that stores data in a BSON format, a dynamic schema document structured like JSON. Mongo hits a sweet spot between the powerful query ability of a relational database and the distributed nature of other databases like Riak or HBase (1). MongoDB was developed keeping in mind the use of the database in a distributed architecture, more specifically a Shared Nothing Architecture, and that is way it can horizontally scale and it supports Master-Slave replication and Sharding. MongoDB uses BSON to store its documents. BSON keep documents in an ordered list of elements, every elements has three components: a field name, a data type and a value. BSON was designed to be efficient in storage space and scan speed which is done for large elements in a BSON document by a prefixed with a length field. All documents must be serialized to BSON before being sent to MongoDB; they're later deserialized from BSON by the driver into the language's native document representation (2).