The combined requirements of a highly competitive gaming market, rapid corporate growth, and ambitious development plans have compelled Goodgame Studio to consolidate several clusters into a single cost-effective and high-performance MapR cluster featuring Hadoop, Apache Spark and Apache Drill. Goodgame Studios leverages Apache Mesos instead of YARN as its general resource management system and relies on Spark as its computational and analytics engine. Additionally, the seamless integration of HP Vertica and Drill into the MapR platform as SQL query engines enables Goodgame Studios to immediately explore data managed in MapR-DB, the NoSQL database.
“From an IT perspective, we wanted the simplest way to maintain the different components such as file systems, databases, and compute engines,” said Andreas Scheel, Senior Data Warehouse Development, Goodgame Studio. “Moving from a multi-cluster Hadoop solution to a single MapR cluster with Hadoop and Spark have enabled us to build a consolidated, co-located data platform with scalable storage as well as the computational engines we wanted without any kind of bottlenecks. This makes for a significantly faster data-to-action cycle.”
With their stated goal of becoming a worldwide leader in gaming, Goodgame Studio’s philosophy of “data always beats opinion” is key to creating highly compelling gaming experiences for their players. By shifting to real-time analytics and recommendations powered by Apache Spark, Goodgame Studios can immediately see customer responses when the company makes changes to its massive multiplayer online (MMO) games.
Switching from a competing Hadoop distribution, Goodgame Studios’ consolidated MapR cluster requires about one third of the hardware than they previously needed, reducing costs and maintenance requirements. The multi-tenant capabilities in MapR combined with its encryption and granular access control features also make it easier for Goodgame Studios to ensure against unauthorised data access.
Furthermore, Goodgame Studios points to the value of the MapR solution in terms of its commitment to open APIs and standards like NFS and POSIX.
“We needed an open platform so we can implement any kind of software and introduce new technology components whenever we wanted to. We could not do this with other Hadoop distributions,” explains Scheel. “With MapR, there are no limits on opportunities. If something comes out that's new, we can use it immediately, we don’t have to wait.”
Goodgame Studios also plans to implement the recently announced in-Hadoop document database capabilities that will soon be available in MapR-DB. This will speed the application development cycle with the easy-to-use JSON paradigm.
“Goodgame Studios is yet another strong example of the value that the MapR platform can bring to an innovative and ambitious organisation,” said Jack Norris, Chief Marketing Officer at MapR Technologies. “Their ability to rapidly develop and securely deploy so many key applications on a single cluster is an excellent demonstration of the power of MapR.”