Letting Go of the Canoe: Embracing a Modern Disaster Recovery Strategy

There is a popular Buddhist parable that involves a person travelling along a path and coming to a great expanse of water. With great effort, the person gathers twigs, grass, and other materials, constructs a canoe, and makes their way safely to the other side.

Once there, the person finds that the canoe is heavy, gets caught on trees, and impedes their progress. However, the person is reluctant to part with the canoe, because they spent so much time and energy constructing it. The vessel served a crucial purpose, and maybe it’ll be useful again.

Yet, to move on effectively, the person must be willing to let go.

Traditional approaches to Disaster Recovery remind me a lot of this story. The term Disaster Recovery tends to immediately conjure up thoughts of replication, failing over, and failing back. But why is that?

I’d argue that it is, in large part, because of an underlying assumption that the storage in question is a relative silo, with limited flexibility. We assume data exists in a single location, so naturally you must copy that data to another location. Then, you have to figure out how to activate that copy upon a trigger and continue business operations using that copy. Finally, at some point, ‘normal’ operations must be restored once the major event has been resolved.

All those steps require planning, implementation, maintenance, overhead, and expense – it’s a complex scenario being managed by staff that is already stretched thin.

Sometimes, doing what’s best for your organization requires ditching that old DR canoe and adopting a more modern approach.

What if we freed ourselves from all those steps, and instead imagined a scenario where our data was stored in a manner that was just as durable and resilient as what we could accomplish on our own? Furthermore, what if it was performantly accessible from virtually anywhere, as if by magic?

A hybrid cloud architecture can deliver this seemingly magical scenario. A hybrid cloud architecture leverages object storage located in hyperscaler datacenters, and places intelligent cache devices at the edge. Data is readily available to end users virtually anywhere, but is ultimately stored and protected in the cloud, allowing customers to take advantage of the durability and resilience of public cloud storage systems. The result is a new storage and DR paradigm that is just as durable and resilient – if not more so – than what individual organizations can effectively build, manage, and maintain on their own.[1]

Disaster Recovery in the Hybrid Cloud Era

Many other resources and costs associated with traditional DR, typically served by third-party products, also diminish Just to name a few:

• Virtual infrastructure requirements are a fraction of what was needed previously
• Traditional backups are replaced by a versioning filesystem with immutable writes
• Bandwidth costs between sites are dramatically reduced because site to site replication is no longer necessary

Disaster Recovery becomes a matter of simply spinning up an edge appliance and pointing it to the object store – a process that takes about 20 minutes.

Corporate sites residing in a natural disaster-prone area are automatically protected – the data is sitting safe and sound elsewhere.

If a utility worker accidentally cuts a corporate site internet connection with a backhoe, or power is lost at a single site – that’s ok. Only one corporate site will be impacted.

If struck by a ransomware attack – that’s ok. With the right hybrid cloud solution, IT can quickly grant secure access to recent versions and minimize business disruption, with a Mean Time To Recovery measured in minutes, not days.

This new architecture does not depend on all sites coming back to a single point of failure on-prem location where the data is stored. The durable, resilient cloud is now the center, where even the lowest levels of resiliency see data sharded across multiple datacenters.

Disaster Recovery with Nasuni

That is how we, at Nasuni, view Disaster Recovery. Cloud object storage allows us to be unencumbered by the shackles of traditional Disaster Recovery and to approach DR in a new way. A much simpler, streamlined, and elegant way.

All the hard parts have been outsourced to entities that specialize in data durability, availability, and resiliency. You no longer need to build, manage, and maintain a system to provide these benefits on your own. They are built into the Nasuni platform. This allows you to focus your efforts on what is really valuable to your organization – not merely maintaining, but using, data to gain critical insights that can help propel your business forward against the competition.

By adopting a modern hybrid cloud architecture, you aren’t re-inventing the wheel. You are simply leveraging resources and technologies that weren’t previously available.

———————————————————————————————————————————————————-

[1] In my 2+ decades of experience, most organizations reach between Tier II and Tier III for their on-prem datacenters, with co-lo’s providing the possibility of Tier IV, but still highly dependent on customer design and implementation within their cage. By contrast, Storage-As-A-Service from hyperscalers start at Tier III in terms of uptime SLAs: https://www.coresite.com/blog/breaking-down-data-center-tiers-classifications