New architectures require new paradigms. Sometimes, scaffolding around older frameworks works. It preserves our investments in the old and leverages our hard-earned skills. With physical infrastructures, the building process is occasionally done very gracefully but it might take them years to be fully appreciated. But ambition and creativity often make us sweep away the old and leave us breathless with the break from the past.
Cloud based architectures have that mix of the old and the new. The reasons for cloud adoption are varied – from agility to economics to freedom from lock-in. To continue reaping the hard-earned benefits of rapid advancement in a young industry, we want to leverage our investments and skills. But the more efficient cloud environment necessitates some changes as our world is being overrun by data exhaust.
In the past, we didn't have a need to figure out trends based on clicks, likes, tweets and mobile texts. Elections didn’t result in massive, and almost instantaneous technological ripples. And while the data stores we once had in place were extremely efficient at working with defined data types and pre-existing relationships, they fell short when it came time for them to discover new patterns and react in near real-time to unstructured data.
The OpenShift team at Red Hat is at the vanguard of the cloud transformation. Much has been written about Platform-as-a-Service (PaaS) redefining the ways applications are developed, deployed, managed and scaled, and the emergence of a devops role that drives this new paradigm forward. Our interactions with developers at the forefront of this trend have reinforced that this new architecture imposes new requirements. Scale-out infrastructure requires a more flexible data management system with characteristics suited to the new cloud paradigm. Enter NoSQL.
The movement towards NoSQL has not gone unnoticed. Examples abound of agile development shops and large enterprises alike adopting NoSQL - in greenfield deployments as well as complementing existing application infrastructure. While there are a variety of NoSQL tools available that differ based on the way they store data (document, key-value, graph, etc), MongoDB has been among the most popular with thousands of downloads and deployments in demanding environments. Analysts have taken notice too. Recently, the JBoss team at Red Hat announced a collaboration to extend the popular Hibernate framework to facilitate NoSQL access for Java developers.
OpenShift PaaS has been providing NoSQL through our support of MongoDB and our relationship with 10gen, the company behind MongoDB, since May 2011. Sharing the principles of open source and community adoption, we have found this to be a very fruitful partnership and of great interest to developer communities.
Today, we advance this partnership with an investment in 10gen. OpenShift users have likely noticed that we recently upgraded to support the most recent MongoDB 2.2 release and added new features. We plan to continue to work together to allow the OpenShift community to take advantage of the latest advancements in MongoDB and NoSQL frameworks in their applications, and to demonstrate this integration to our communities globally. And we expect the benefits of our continued collaboration with 10gen to be felt by the users of all our platforms - OpenShift Origin community, OpenShift Online service and OpenShift Enterprise - enabling public, private and hybrid PaaS deployments.