3 Best Practices for Optimizing Your OpenStack Backup and Restore Operations

In the years since the emergence of cloud infrastructure, native workload backup and recovery has become an increasingly pressing issue. Public and private clouds are now pervasive in modern IT stacks, making efficient workload backup and restore operations a necessity in order to safeguard against extreme data loss (disasters) as well as everyday data loss and mobility issues (migration between tenant workload spaces, upgrades, etc).

Here are a couple of simple ways to ensure that your OpenStack backup and restore operations are as effective as possible.

3 Steps to Optimizing Your OpenStack Backup and Restore Operations

The promise of a tenant-based, highly scalable cloud framework is what makes OpenStack so appealing. Businesses can now leverage these types of environments to run high-performance computing (HPC) workloads on existing hardware, including traditional and next-gen VMs, and other production workloads.

Due to the critical nature of this data, OpenStack backup must be made a core part of your business’ continuity and disaster recovery infrastructure and needs to function seamlessly since it may be needed at a moment’s notice.

However, the implementation of robust backup and restore operations has become more challenging as organizations transition to highly distributed cloud and virtual environments. Optimization of these functions requires foresight, planning, and diligence.

1. Review OpenStack Documentation

Before putting backup and recovery procedures in place, cloud administrators should thoroughly review the OpenStack documentation of the release they are rolling out. The documentation will outline up-to-date best practices on protecting the applications and configuration files underlying the OS components (though they often lack necessary details regarding the block storage and object storage repositories).

It’s essential that cloud administrators are given the ability to perform regular backups of OpenStack services, network information, configuration files, databases, and VMs. VM backups can be challenging when leveraging scripting methods, and are much easier with a cloud-native, snapshot-based backup solution. Capturing this workload data becomes invaluable when you need to restore VMs, volumes, specific files, and full project workloads: it’s a one- or two-click recovery rather than a complex process of manually stitching everything back together.

2. Deploy High Availability

Disaster recovery and high availability facilitate business continuity in the event of a failure. Large-scale failures with respect to disaster recovery are focused on individual component failures. Most enterprises deploy OpenStack using a controller and compute layout where the controller runs nova-scheduler and other management services.

While a failure of the controller won’t disrupt already running workloads on OpenStack, it’s essential to maintain 99.99% uptime for the control plane. As such, the controller must be deployed in highly available configurations so that applications can scale as needed and OpenStack services remain accessible at all times.

Any OpenStack disaster recovery plan must:

• Ensure workload data replication through backup/restore, application-level replication, or storage replication
• Ensure availability of VM images needed to run hosted workloads on the target cloud
• Capture the cloud management stack metadata that’s relevant for protected resources/workloads, either as continuous replication of the metadata or point-in-time snapshots

3. Implement a Native Data Protection Solution

Stein, the latest OpenStack release, includes Karbor, a data protection orchestration service that implements libraries and services for providing project-aware data protection orchestration for existing vendor solutions. But Karbor is complex, offers no support model, and demands a great deal of command-line work in order to capture and recover workloads on demand. Consequently, it’s challenging to role out tenant-driven data protection on a broader scale.

TrilioVault, by contrast, enables tenant-based backup and recovery operations from the existing Horizon GUI… so cloud users can define data protection policies that are appropriate for their workloads. This drastically reduces administrative overhead.

TrilioVault also captures complete workloads, including: applications, OS, network configurations, storage volumes, security groups, and of course, the VMs themselves. This gives administrators the ability restore their OpenStack projects as broadly or as granularly as needed — without the complex re-building processes.

Maintaining Best Practices for Cloud Backup

As more enterprises continue to adopt OpenStack cloud for production purposes, the platform is increasingly used to support a mix of IT workloads, including stateful and stateless applications. This makes data protection (and by extension, disaster recovery) a priority.

By leveraging Trilio’s robust data protection solution, cloud administrators can readily replicate and restore workload data, including applications, metadata and data, and selectively restore missing or corrupted files.

Connect with a team member today to learn how Trilio can help with your backup needs.