Diego Ferreira

Switching from MongoDB to DataStax Enterprise for High Performance – Interview with Early Birds

By Diego FerreiraDecember 18, 2015

This post is one in a series of quick-hit interviews with companies using Apache Cassandra™ and/or DataStax Enterprise (DSE) for key parts of their business. For this interview, we talked with Nicolas Mathon, CTO at Early Birds.

DataStax: Hi Nicolas, many thanks for the opportunity of this interview. Could you please tell us a bit about Early Birds and what exactly you offer?
Early Birds: Sure. Based out of Paris in France, Early Birds was founded 4 years ago, and provides a personalization platform to eCommerce websites. We identify, in real time, the best product or content to display to each user. We are actually a French leader of product recommendation. We currently deliver personalized customer experiences to French top retailers and aim to internationalize next year.

DataStax: What makes your recommendation engine successful, what differentiates you to other similar applications? What makes you excel at customizing the user’s experience and increases key metrics for businesses?
Early Birds: First of all, we are a true real-time platform. We learn from users browsing behavior in few tens of milliseconds and we are able to personalize customer experiences, even from anonymous users, just after the first interaction.
Secondly, we are the only platform which provides a new experience to website users. Thanks to our interactive recommendations, each user can browse products with a like/dislike system. We also offer an online predictive personal shopper which aims to reproduce the behavior of a seller in store.

DataStax: Did you use a different technology before you started using Cassandra?
Early Birds: Yes, we started with MongoDB and Redis.

DataStax: Why did you decide to use Cassandra? What kind of data is stored there?
Early Birds:  We work with the top French eCommerce websites so we drive a lot of traffic. We store many kinds of data but essentially profiles information, interactions between users and contents or products, and point of interest. After a few months, we started to have performance problems with MongoDB, essentially for writing hundreds of data by second. MongoDB is good for storing unstructured data and querying them. But it’s not the best solution when you store several GB of data each day.
So we benchmarked several systems and decided to mainly switch to Cassandra. We kept MongoDB though. We use MapReduce with Spark on Cassandra and store results in MongoDB. So we can use MongoDB for complex data querying and the power of Cassandra for storing a large amount of data in real time. Since we switched from only using MongoDB to Cassandra, we eliminated performance problems and reduced our API response time.

DataStax: How would you sum up the benefits you’ve achieved with DataStax Enterprise (DSE)?
Early Birds:  The database performance and high availability are definitely a win for us.

DataStax: What caused you to use DSE over open source Cassandra?
Early Birds:  We wanted to start with something easy to deploy and use, bundled with the good tools. Now, DSE’s tools like DevCenter and OpsCenter became our daily tools and we can’t imagine working on Cassandra without those. We use Datastax OpsCenter to monitor our clusters and to deploy new machines. We also use Management Services (mainly backup and repair services). And our developers use DevCenter as a fast querying tool.
Furthermore, we wanted to have good support overall and DataStax reassured us tremendously.

DataStax: Tell us about the future of your project, do you intend to leverage other parts of DSE to make it a reality?
Early Birds: DataStax is crucial for our core businesses, which is our data platform and our recommendation engine. So in the near future, we want to go further and also provide powerful analytics and Business Intelligence tools to our clients.

DataStax: What advice would you give to other startups that are thinking about using Cassandra for the first time in their solutions?
Early Birds: Keep it simple. Start with only the tools you really need. DataStax offers a really easy-to-use system and you can start without installing or using everything.
Regarding Cassandra, you have to think about your schema and how you will query it. It’s a really different way of thinking than with other DBMS. And it’s not exactly simple to change your schema afterwards. When you know precisely what you want to do, and when you designed it well, it’s a really powerful system. Also, don’t hesitate to ask the DataStax team; they always reply really fast!





SHARE THIS PAGE
SUBSCRIBE

Comments

Your email address will not be published. Required fields are marked *

Tel. +1 (408) 933-3120 sales@datastax.com Offices France Germany

DataStax Enterprise is powered by the best distribution of Apache Cassandra™.

© 2017 DataStax, All Rights Reserved. DataStax, Titan, and TitanDB are registered trademark of DataStax, Inc. and its subsidiaries in the United States and/or other countries.
Apache Cassandra, Apache, Tomcat, Lucene, Solr, Hadoop, Spark, TinkerPop, and Cassandra are trademarks of the Apache Software Foundation or its subsidiaries in Canada, the United States and/or other countries.