MongoDB is being used to log a high volume of user activity data and perform data analytics in the retail space. The system requires handling hundreds of thousands of writes per second, flexible schema, fast querying and sorting on varied fields, easy deletes of old data, and a general data model that captures attributes such as user ID, device information, item ID, and location. To meet these requirements, MongoDB is used with the High Volume Data Feed (HVDF) framework, which allows for sequential writes, flexible schema, fast querying, easy deletes, and time partitioning. HVDF enables the creation of separate collections for each time interval, making it easier to query data within a specific time frame. The system also uses map-reduce to perform analytics tasks such as calculating the number of occurrences of each item pair, grouping the most popular pairs in a separate collection, and creating secondary indexes on item IDs and their counts. Additionally, MongoDB is integrated with Hadoop for complex number crunching of massive datasets. Overall, MongoDB provides a scalable solution for performing user analytics and gaining valuable insights into customer behavior.