The rise in containerized workloads brings with it the need to provide back up and disaster recovery. Afi, a cloud data management firm, threw its hat in the ring today and launched its namesake cloud-based replication and resiliency Data Platform for containerized applications and services.

The new service monitors, protects, and recovers Kubernetes-based workloads running in Microsoft Azure or Google Cloud Platform, which CPO Vincent Marino says differentiates Afi from legacy services. This market includes traditional backup vendors such as NetBackup, Commvault, and Veeam, and rising hotshots like Rubrik or Cohesity.

As organizations continue making accelerated leaps from largely on-premises environments to the cloud, systems administrators are turning their attention toward backups. “We're seeing fast container adoption — as much as 5% — of enterprise servers hosting container applications, and their share is growing. At the same time, the number of solutions enabling native Kubernetes backup is very limited,” Marino said. 

Afi tailored its platform to meet the growing container-based enterprise workload market that the IDC forecasts will grow at 163% compound annual growth rate (CAGR) from 2020 to 2023 and exceed $1.9 billion, with more than a quarter of them used as stateful containers.

The company released its first software-as-a-service (SaaS) for backup and migration in 2017. 

The Afi Data Platform uses artificial intelligence (AI), which Marino said automatically identifies application components to distinguish between different types of container workloads — think: application servers or load balancers — and detects stateful containers. Ai-based application monitoring and downtime prediction also helps to automate the disaster recovery process, he added. 

It first scans application components, identifying storage resources, application pods, namespaces, and Kubernetes cluster configuration. Once the AI-based engine identifies the container type, Afi then uses this knowledge to create an equivalent Kubernetes configuration and replicate the primary instance to a secondary location. According to the vendor, the platform can maintain a replication delay of under 15 seconds, using tools native to Azure and Google Cloud combined with proprietary in-line deduplication.  

After the replication application is established and ready for action, attention then turns back to the primary application where the AI-based engine monitors for abnormalities. The end game is to reduce the time required to configure and manage backups. 

“Organizations worldwide are reconsidering the importance of their IT systems resiliency and the ability to instantly scale to accommodate higher load,” said Marino. “In such an environment, Kubernetes applications are favored over legacy monolithic systems, as they scale natively. Afi Data Platform helps to implement a reliable disaster recovery strategy for Kubernetes applications, ensuring their resiliency.”

Containers were initially viewed as a perfect match for stateless applications that did not require stored data to operate or support a running application. Their marital success has since lured the attention of organizations that seek to allow containers to tap into a central data repository and support stateful applications.

The challenge in container storage is meeting applications performance requirements that are free from the burden of physical servers – a feat that is far easier said than done. 

That said, no matter the unrelenting pestery of simplifying and securing container storage, vendors young and old have been spinning out updates to milk this market opportunity for all its worth. 

Diamanti went deep with the latest update of Spektra last week with a ready-made appliance that hosts containers, and more importantly, handles the networking and storage requirements for them — which is a particularly handsome service for organizations that want to run container workloads on-premises in their own data centers.     

NetApp launched Project Astra, an enterprise-class storage and data services platform for Kubernetes that enables both application and data portability. Rancher Labs’ Longhorn was made available in June as a cloud-native container storage service to enable vendor-neutral persistent storage that supports stateful workloads on Kubernetes clusters.