It’s no secret that Hadoop and NoSQL databases like Cassandra continue to gain popularity, if the technology press and media headlines are any indication. Hadoop has been a darling ever since the term Big Data was coined. But earlier this year Gartner dropped some ice water on the party in their report “Gartner Survey Highlights Challenges to Hadoop Adoption”. Stating “Survey Findings Reveal That More Than Half of Respondents Have No Current Plans for Hadoop Investment.”
One of the problems no doubt is the challenge contrasting and figuring out exactly what type of NoSQL database is most appropriate for a particular project or initiative. Should we abandon relational databases all together? There are after all, literally dozens of databases to choose from (see the 451Research Oct 2014 Data Platforms map). You can request the latest June 2015 edition of the database map from 451Research here.
Like Apple (see The Abandoned Facebook Tech That Now Helps Power Apple and Apple’s secret NoSQL sauce includes a hefty dose of Cassandra) and many others, we at Reltio selected Cassandra as our NoSQL database.
As I detailed in the Cassandra meetup back in March, we’ve added a healthy dose of our secret sauce on top to create what we call the Commercial Graph. The inspiration for our graph technology comes from consumer data-driven applications such as LinkedIn (economic graph), Facebook (social graph) and Google (knowledge graph). At Reltio, we believe that the only way to deliver infinitely scalable, flexible and extensible enterprise-class data-driven applications is to blend the power of C* together with graph technology. The days of single persistence layers have been replaced by polyglot strategies, and also the rise of multi-model databases. I’ll discuss these topics in more detail in a future blog post.
We have and continue to combine and integrate the best technologies into a modern data management platform will allow enterprises to handle any scenario, and deploy the latest generation of data-driven applications. Our goal is to democratize access to data for frontline business users while giving IT the technology they need to ensure reliable data, and to be agile and flexible to the needs of the enterprise.
The reason we decided to build our own, rather than use an off-the-shelf graph database was that popular graph databases such as Neo4J provide flexibility, but they didn’t have the horizontally scalable capabilities we needed. Our modern data management hybrid columnar and graph store, gives us the flexibility of schema-on-read, graph relationship modeling, combined with infinite horizontal scalability across entities and limitless attributes.
So like LinkedIn, Facebook and Apple, we rolled our own to support big data volumes and multitenant cloud. We have since been able to easily incorporate Apache Spark Graph analytics and machine learning, through the use of multiple network graph representations out of a single persistent data structure. Allowing us to support causal impact analytics, non-obvious entity analytics inference, to offer a powerful framework for predictive insights to address areas such as fraud detection and more, all on the same pool of data.
It’s one of the reasons we have been able to grow so quickly, delivering data-driven applications with that offer both operational execution, and analytics to solve any business problem, for any industry. Our horizontal use cases range from replacing legacy master data management tools, Key Account Management applications, to M&A, data monetization, and even compliance. And we have very quickly addressed some of the biggest unmet needs in the highly regulated, healthcare and life sciences industry.
Our philosophy has certainly been well received. I was honored to speak at the 2015 DataStax Cassandra Summit on the topic of “Powering Data-driven Applications with Cassandra”. The session was well received with great interaction with the audience as I provided a glimpse into our capabilities and architecture.
We also sponsored and exhibited at the event, and were delighted when Billy Bosworth, DataStax CEO stopped by to chat. He called us “veterans” given that Reltio had one of the most long standing (since 2011) uses of Cassandra in the enterprise. We have a long history with DataStax, sharing the same VC, Crosslink Ventures, and have received incredible support from their team, as we have grown together. Last Spring co-founder Matt Pfeil provided a very nice endorsement of our series A funding.
As technology continues to evolve, and more and more database options appear on the market, we guarantee that our customers will continue to benefit from innovations such as Cassandra, and whatever comes next.
Please contact me if you have any questions or if you would like a copy of my presentation.