This post was published on Hortonworks.com before the merger with Cloudera. Some links, resources, or references may no longer be valid.
As modern data architecture has developed over the years, it’s becoming increasingly important for businesses to identify ways to better manage their data and eliminate data silos. The new Open Hybrid Architecture Initiative aims to help businesses do just that by building an enterprise deployment model that’s reliable, seamless, and able to run across on-premises, multi-cloud, and edge architectures. The initiative will add value to customers by cutting costs and breaking down data silos, and will help create a reliable model with consistent security and architecture. Here’s how.
The Need for the Initiative
The Open Hybrid Architecture Initiative acknowledges that because today’s data comes at us in volumes and speeds with which previous infrastructures simply cannot cope, we must adjust how we approach data storage, access, analytics, and governance. Infrastructures of the past were often expensive and wound up creating silos where some data existed here, other data existed there, and there wasn’t a cohesive way to manage the life cycle of all of that data.
The sheer quantity of data we deal with these days necessitates less expensive, ever-expanding storage and the ability to spin up resources as they’re needed and spin them down when they’re no longer required. The Open Hybrid Architecture Initiative aims to incorporate the best parts of the concept of the cloud while providing a framework for operating locally on premises as well. It’s about bringing cloud-native technologies—and the ethos of operating across a huge, always-available platform—down to your local data center, making your on-premises architecture and your cloud architecture operate in an identical way.
On-premises Hadoop solutions are ready to go: they’re able to understand and return answers to queries from users as soon as they’re implemented. Currently, cloud data architecture requires setting up virtual machines, provisioning storage, installing Hadoop, setting up Spark clusters, and maintaining query instances, among other procedures. This needs to change, which is part of what the Open Hybrid Architecture Initiative is all about. The use of a data plane allows for seamless movement and interaction with data—whether it’s in the cloud or your data centers—that’s supported by open standards and industry leaders.
What Does This Mean for Your Data?
There are three foundational principles for this initiative:
- Reliability. The cloud-like nature of the architecture has redundancy, security, and isolation built in, along with the tools necessary to orchestrate reliability and high availability across the entire installation.
- Uniform security. A piece of data that comes in needs to be tagged, realized, and governed, regardless of where it currently lives, where it first lived, or if it’s been destroyed.
- Consistent architecture. This might be the most important part of this entire concept: the architecture of how you manage and interact with your data needs to be the same whether it is running on premises through servers, or up in the cloud. It should look identical and be portable between running locations. The initiative makes real progress toward this end.
Exciting times are ahead with this new initiative to bring cloud-native technology to the enterprise data center and a redesign to modern data architecture.
To learn more about the Open Hybrid Architecture Initiative, click here.