Simply backing up data is not enough to protect data and ensure business continuity – particularly in an increasingly competitive, complex and risky environment.
This emerged during a webinar hosted by Metrofile Cloud, in which backup and disaster recovery (DR) experts outlined the differences between backup and DR, common DR models and why DR as a service (DRaaS) can prove to be the ideal option for business.
Polls of attendees found that 56% have only backup in place, 2% have DR only, 4% have neither backup nor DR in place, and 36% have both backup and DR. On the question of how comfortable they are with their DR plans, 26% are very confident, 37% are somewhat confident, and the rest are either not very confident or not at all confident.
Differentiating between backup and DR
Trent Lester, Head of Technology at Metrofile Cloud, noted: “Backup and DR are quite different, and ideally, organisations should have both in place. Backup is for data protection – for example, if you need to recover a file that was deleted accidentally, if you have a ransomware attack, and for compliance. I see backup as the last line of defence in terms of layering your security. On the other hand, DR is for business continuity. You can simply backup data and applications to the cloud, but for DR, you also need an infrastructure for where to restore that data, with all the configurations and interconnects required to run your business.”
DR options
Nicholas Keene, Solutions Architect at Metrofile Cloud, outlined three common DR models. “In the simplest option, critical data is regularly backed up to cloud storage using the 3-2-1 rule, with granular recovery options. This is a lower cost solution that is simple to deploy and easy to scale. However, it has longer recovery times and few automation options, and is difficult to grow into a full DR solution at a later stage,” he said.
Another more complex option is a periodic replication solution that leverages the backups for DR, taking snapshots of systems and storing them in a backup location and then, at regular intervals, doing automatic restores to the DR infrastructure.
Keene said: “This allows for a much faster recovery time with the benefit of being off site. It also allows for more automation and testing. Because it leverages the backups, it saves on storage costs and reduces complexity in terms of deployment. It offers lower complexity DR and supports heterogeneous environments. However, this model can be more expensive than simple backup storage, and some manual action is required. This solution suits mid to large organisations with moderate recovery time objectives (RTOs).”
A third option is an active standby solution that offers the lowest recovery point objectives (RPOs) and RTOs for the most critical systems.
Keene said: “This option can replicate in near real-time directly from the hypervisor. It supports complex environments and automation requirements, and is ideal for organisations with critical workloads and short RTOs. It offers near real-time recoveries and is highly customisable with excellent orchestration options to fully automate failovers. However, it is the most expensive and complex option with the highest management overhead, and it typically works with enterprise hypervisor platforms only. In addition, many of these solutions offer little to no support for physical systems.”
To choose the right solution, organisations should ask what their acceptable RTOs and RPOs are, what workloads need protection, what the budget is, what regulatory requirements must be complied with, what level of automation is required and whether they have the right expertise internally, Keene advised.
Cloud DRaaS can reduce the cost, complexity and stress associated with DR, he said. “With cloud DRaaS, you leverage the service provider’s scale and expertise, and don’t have to go out and buy the infrastructure yourself. They are secure, hosted in a highly redundant data centre and should have security and compliance certifications in place, which ensures your systems are secure. It is also scalable and automated, and when you do have a disaster, you have the assurance the DR will work.”
Lester also highlighted Simple Storage Service (S3), which ties into backups to tier storage. He said: “Organisations may want to keep data locally for two to three days, then move older data to the cloud, then archive data to S3 storage. Think of S3 as a giant online storage unit with buckets or folders containing data or objects. S3 is super reliable, built with considerable redundancy and uptime in mind. It comes in very affordable packages and is available in a pay-as-you-go model.”
Share