It seems to me that backup and recovery or BDR (Backup and Disaster Recovery) for short has always stuck in the past, utterly unaffected by the tectonic changes happening in the rest of the industry. Most of the solutions were designed during the 1980s using client/server technology and are architected from a storage rather than an application point of view. Those architectures worked well when the applications were static and once deployed, it remains unchanged for years, so backups not worrying about the application configuration made sense then. For example, how often would you reconfigure your Oracle RAC server?
Fast forward , and we live in a brave new world. Applications are a lot more dynamic and cloud agnostic. They dynamically scale, and users deploy and manage them using DevOps/Platform engineering. The old way of backing up a few disks does not guarantee a quick recovery of your business-critical applications.
At Trilio, we redefined the Backup and Disaster Recovery (BDR) paradigm with an applications first strategy . We embarked on our journey with a proper native backup and recovery solution for OpenStack clouds. It provided a simple-to-use UI for tenants who do not have to be backup experts and navigate the complex backup infrastructure like master, media, and cloud servers, deduplication policies, complex retention policy, and the whole shebang. The tenant-driven UI puts the applications in the driver's seat, and users can set up backup policies without worrying about the application complexities. We preserved the same ease of use whether the application is a 30-node Kafka cluster with complex local and public interconnectivity, security groups, metadata, and storage. With a click of a button, the user can protect these applications, and with a click of a button, the user can restore the entire application, including public IP addresses. It took the guesswork out and provided a better RTO solution than any other solution.
We extended the same paradigm to the cloud-native world. The recently awarded patent scalable cloud-based backup method is a testament to our commitment to the application and user-centric backup and recovery approach.
"A computer-implemented system and method of backing up and restoring a containerized application or a cloud-based application using a datamover service includes determining a stateful set of services of the containerized application or cloud-based application to be backed up. A persistent volume associated with the determined stateful set of services of the containerized application or cloud-based application is identified. Then, a snapshot of the identified persistent volume is created and a new persistent volume is created from the snapshot. The created new persistent volume is attached to a datamover service. Data from the created new persistent volume is then copied to a network file system or storage system using the datamover service, thereby creating backup data stored in a storage system."
As the patent abstract describes, our Kubernetes solution is built as a cloud-native solution that scales with your platform. Deploying and managing Trilio is no different than deploying any other cloud-native application and is well integrated with the Kubernetes ecosystem. Users can use cloud-native tools such as Prometheus, Grafana, Kyverno, KubeVirt, Red Hat ACM, Operators, Ansible, Kubectl, ArgoCD, and GitOps to manage Trilio backups and recovery. Trilio supports all application types, including labels, namespaces, operators, and Helm releases, so users are not constrained by how they deploy their applications. Trilio's backups are industry-standard qcow2 images that are space efficient and do not require expensive and unwieldy media servers and deduplication infrastructure. Trilio supports NFS and S3-compatible storage, so your backups are not locked down to clunky hardware in your data centers but can migrate between clouds.
The new generation of applications needs new data protection solutions that are application-aware, highly scalable, and multi-cloud friendly that lend themselves to the DevOps approach to life cycle management. Older solutions offer stiff resistance to your DevOps, stifling innovation and draining the efficiencies new trends promise to bring. As with your applications, your data protection solutions must be cloud-native too.