We are embarking on a new initiative and wanted to get community inputs if using Janus Graph is a good fit. Any alternative suggestions also welcome.  

• New records are added regularly. Assume every visitor to the website
• A Visitor can become a customer
• Customer can join a group voluntarily. A group customer joins could be high cardinality or low cardinality. 
• Customer will be added to group by the system based on characteristics. (For example Age Groups, Male/Female/Other, Country etc...)
• Customers can move among groups or cease to be part of the group
• Customers would be part of a group for a given duration. For example they are part of a group when an event is happening.
• Customer has multiple Unique identifiers to get them by that key (could be customerId, subscriptionId etc)

We are looking at 300 - 400 million entries.

We are expecting decent amount of OLAP requests like :

• Give me all the customers that belong to a group
• Give me all customers that belonged to a group but not any more.
• Give me all customers that a group but also belong to another group.
• Give me related customers (referrals)

We have Elastic Search, Cassandra  and others being used in our ecosystem

thank you,
BG

Hi Basanth Gowda,

The fit between your use case and janusgraph does not seem particularly good. The mean reasons for my opinion are:

1. your datamodel still seems rather simple (website visitors and groups); it could easily be handled in a relational model using SQL
2. analytical queries are important to you, while this aspect of janusgraph is still in its infancy (you might want to check an OLAP meeting to be planned)

I am curious to see whether other opinions pop up!

Best wishes,    Marc