As a social network with increasing popularity, WeSpeke began to feel some growing pains with the large amounts of data they were now handling. Their NoSQL database service was starting to get throttled, impacting the end-user experience. The huge amounts of data they were starting to collect became hard to backup and manipulate. Even reporting tasks became hard to manage. Due to its excellent capabilities of handling large amounts of data that is highly interrelated, a graph database seemed like a natural fit for WeSpeke, and they chose a Neo4j solution with GrapheneDB.
Prototyping to stay on-track
Working with deeply associated data is much easier in a graph database like Neo4j, as it is much faster to traverse the data. Being new to Neo4j at the time, WeSpeke needed a way to prototype quickly and ensure queries were on the right track. Through the easy-to-use GrapheneDB dashboard, WeSpeke was able to provision test environments quickly in order to prototype before moving to production. This provisioning is so easy, WeSpeke uses 3 different environments in addition to production.
After moving to GrapheneDB, the WeSpeke team has found it easier to manage the large amounts of data they deal with on a day-to-day basis. The automated nightly backups have been especially helpful for WeSpeke. Self-hosted Community Edition Neo4j solutions do not come with automated hot backups by default, but this is something GrapheneDB offers as an added value. Not having to worry about managing the database has made it easier for the WeSpeak team to focus on development.
While choosing Neo4j was a natural fit for WeSpeke, laying all of the groundwork necessary to scale on your own can be costly for a startup. Partnering with GrapheneDB resulted in cost savings for WeSpeke, as they didn’t have to worry about spending time or money spinning up machines. Since it’s so easy to spin up or down instances with GrapheneDB, and clients only pay for what they use, the cost of running the databases the product needs is more manageable.
While the WeSpeke team is currently still in the process of fully migrating all of their data to GrapheneDB, they have experienced a 3x to 4x increase in performance out of the queries running on their Neo4j instances, making their whole system run faster.
Not only is WeSpeke’s overall system faster, but GrapheneDB’s response time to questions from the WeSpeke team are always extremely fast.