MongoDB - a beginners review
June 03, 2016
MongoDB - a beginners review
Score 5 out of 10
Vetted Review
Verified User
Overall Satisfaction with MongoDB
MongoDB was used to store middleware transaction messages. These JSON messages were sent across various heterogeneous systems like Maximo and Click Sycleo etc. It was only used to store the important messages for local use and references. The main system software IBM Maximo used a relational database. MongoDB was locally set up to store templates of frequent and different types of middleware messages.
- Since Mongo DB stores unstructured data, it was easy to directly store messages in it. We did not have to de-serialize serialize the json data
- Of bulk messages where exported from the Oracle Service bus which was the middleware into csv. The import of all the messages from the CSV and MongoDB was done using a single command
- Complex queries can get confusing at times when comparing multiple constraints. The query language needs a bit of improvement.
- Joins like mechanisms are missing in MongoDB.
- It made archiving of messages easier. You can export from Oracle service bus as CSV and import the message in MongoDB.
Relational DB are not efficient when storing data structure like JSON. Different data structure can be stored without defining the schema. Most relational db might store data like Json as blobs. One single entry would store the entire JSON as blob and you can't query the structure to find out the various data in JSON. You would have to retrieve the whole structure and parse it. Or parse the whole JSON and store the appropriate value of key to a corresponding column in the DB. In MongoDB you can query the single document and also fetch the data of all key value pairs and also its nested key value pairs directly.