Persistent data implications for apps and microservices

Speed and agility are the name of the game, whether you are running track in a triathlon, racing to find cures to the world’s most nefarious diseases, or developing new applications that are changing the way society interacts. Application development teams can have a profound effect, not only on their organizations’ ability to differentiate themselves, but also the world we live in. 

Why are organizations employing containers?

While just a few years ago, some organizations were still concerned with the viability of running production workloads in containers, the benefits of capitalizing on faster development cycles has garnered favor among developers. And, with enterprise-class enhancements delivered by platforms such as Red Hat OpenShift Container Platform, containers have grown from nifty developer projects, to scalable, more manageable infrastructure environments that enable DevOps for the hybrid cloud. 

A challenge posed to application architects is how to enable new applications and microservices across these varied infrastructures and creating and acting on the data they produce pursuant to governance and security requirements. 

Persistent storage for stateful applications

While many application development projects running in containers  just a few years ago were stateless, this new agile development platform has inspired stateful application workloads to be developed in containers. As such, the move to stateful, cloud-native applications is driving new requirements for reliable persistent storage. Applications need a place to store increasingly large amounts of data, but organizations demand performance and portability along with enterprise-class features to support business continuity. Given the importance of storage to application performance and availability, application architects, developers, and engineers need the ability to quickly adopt an integrated  cloud storage solution that works best for their applications.

Container storage is important for stateful applications. It provides a storage abstraction layer for diverse workloads, effectively isolating applications from underlying storage technology, and can provide:

• Performance advantages over available native cloud storage.

• Data portability, and the ability to rapidly evaluate underlying cloud provider instances and storage technology for performance and cost effectiveness.

• Faster failover and recovery than native cloud storage along with more flexible failover (e.g., failover across Availability Zones (AZs)).

But in the end, an infrastructure platform is only as good as the results it is able to deliver. To better understand performance and availability issues around persistent cloud storage, Red Hat engineers evaluated Red Hat OpenShift Container Platform with Red Hat OpenShift Container Storage across several diverse workloads.

Stateful applications make use of many underlying services such as CI/CD pipeline utilities, SQL databases, web-based applications, and NoSQL databases just to name a few. As such, we embarked on testing workloads including MySQL, Jenkins, and MongoDB, to measure the effectiveness of stateful applications running in containers. 

In some cases, performance was increased over standard cloud storage options. In others, Red Hat OpenShift Container Storage allowed rapid evaluation of underlying storage technology to determine the best mix of performance and cost effectiveness. Distinct failover advantages were seen with the ability to fail over more quickly, or to provide failover across Availability Zones to help ensure business continuity. 

To hear more about these results, as well as where the future of container storage and multi-cloud federation is going, please check out the recording of our virtual event: Develop. Deploy. Deliver continuously from October 10, 2019. Register today!