The Library Never Closes: Inside Nasuni’s Always-On Data Architecture

A few weeks ago, during a conversation with a prospect, I was asked a seemingly innocuous but critical question: “What happens to the user experience during a regional power failure if Nasuni is part of our environment?” It’s a great question, one that touches on the core of resiliency and business continuity in IT infrastructure.

Rather than diving straight into technical jargon, I responded with an analogy that came to me in the moment: Think of it like a library system.

At first, it was just a helpful way to explain a complex architectural concept in more relatable terms. But the more I thought about it, the more I realized how well the analogy holds up — not just in describing how Nasuni works during outages, but in illustrating the entire design philosophy behind our global file system. Libraries, after all, are structured, resilient, and designed for access and availability, all values that align perfectly with how Nasuni operates. That realization inspired me to expand on the idea here, through this blog post. Let’s dive in.

How Normal Operations Work

Imagine a large city, and within that city, a central library that houses every title in the system. Outside of the city there are suburbs. Each one of these suburbs has one or more branch libraries; they don’t have copies of all the titles, but they do have copies of the most frequently requested titles for that neighborhood. In this scenario, the central library represents the golden copy of data on object storage, while the branch libraries represent the Nasuni Edge Cache appliances.

Under normal operations, one goes to their local branch and requests some books. Chances are, most of the requested titles are available at that branch and can be taken home immediately. Occasionally, a title might not be available, and in that case, the librarian submits a request to the main library. Leveraging the intra-library delivery system, that title is then available for pickup the next day.

This is how normal operation works in Nasuni: the central object store contains all the data, and the local Edge Cache appliance can fulfill data requests 98% of the time. For the remaining 2%, a request is made to the central store, resulting in a small delay to retrieve the data—typically governed by the bandwidth available to the remote site.

When Your Local Branch is Unavailable

Now, imagine the local branch is unavailable because it’s in a flood zone and the nearby river has flooded. The reader goes to another branch (which they can do, since they’re a card-holding member of the library system) and finds that while it has access to the same overall set of titles, there are some differences in what’s immediately available. Perhaps the people in that neighborhood really like Stephen King, but our reader comes from a neighborhood full of families with kids, so J.K. Rowling is more popular in the now-flooded branch. Fortunately, the intra-library system is still functioning, and the central library — with all the titles — is still accessible, so our reader can check out everything they need.

This is what happens when a Nasuni Edge Cache appliance is unavailable. An authenticated user can access a remote Edge Cache if their local appliance is down and still retrieve their data. It may take slightly longer than accessing their primary appliance due to differences in the data cached at each location, but the data remains accessible.

Additional “Library” Benefits

While the above paints the picture of a single event (the loss of an edge cache), the value of this IT architecture has far-reaching implications for other scenarios as well.

For example, with data potentially available from all branch offices (Edge Cache appliances), the criticality of each individual branch being highly available is minimized, since users can simply reroute to another one. This high-availability model is far simpler and cost-effective compared to traditional approaches that involve replication, failover procedures, questionable failback processes, and “cross your fingers and hope” strategies during an actual emergency.

Another example: the central storage, based on durable object storage that leverages versioning and immutable writes, is analogous to our central library being fully equipped with disaster protection. This includes backup generators, flood mitigation, fire-proof construction materials, and structural integrity capable of withstanding powerful earthquakes. In other words, it’s the equivalent of a Tier IV data center while most organizations build infrastructure in facilities that equate to Tier II or perhaps Tier III standards. The key advantage here is that traditional third-party backup, disaster recovery, and replication solutions become unnecessary, reducing both subscription costs and administrative overhead.

There are multiple ways to implement a global file system, but this is where Nasuni’s hub-and-spoke model really shines. You’re never locked into a single location, and any authenticated user can access their data from any branch.

Ways to Minimize the Impact

As is usually the case, minimizing impact ultimately comes down to how much an organization is willing to spend, whether on infrastructure (when dealing with on-prem environments), or in the cloud. The Nasuni subscription simply provides the intra-library organization and delivery system between the central library and the branch offices. How an organization chooses to set up those library locations — the number of branches, their size, and their placement, all of which carry cost implications — is entirely up to them and based on the needs of the business and its workloads.

Your Data, Available Anywhere

By reimagining enterprise file storage as a modern library system, it becomes easier to grasp how Nasuni delivers both performance and resilience. No matter where your teams are located — or even in the face of outages — your files, like titles in a well-managed library network, are never out of reach.

This IT architecture ensures seamless access, minimizes disruption, and reinforces the idea that availability shouldn’t be tied to a single location, but built into the system by design. And that is exactly why with Nasuni, your data is always open, always shelved, and always ready for checkout.

Tech: Distilled gets to the heart of today’s file data challenges without the fluff. In this series, Ryan Miller, Senior Solutions Architect at Nasuni, unpacks complex technical concepts with sharp insight and real-world relevance. From data security and file locking to the building blocks of a unified file data platform, it’s the kind of practical knowledge that sticks. If you want your tech smart, clear, and just a little bold, this series is for you.