Rethinking Redundancy: How Global File Systems Challenge the Need for Deduplication
Ryan Miller discusses how a global file system like Nasuni helps enterprises better store, protect, and provide access to their data.
January 30, 2025 | Ryan Miller

Deduplication.
Rarely do I engage with a customer who fails to bring up that term. When data growth is explosive, companies look at every available option to optimize storage use. Historically, reducing duplicates of blocks and/or files has been one of the best ways to maximize the ROI on storage costs.
But it is important to step back and evaluate the real problem we’re trying to solve. What is the cause of duplication in the first place, and given advancements in storage, is there a better way?
A Very Brief History of Deduplication
It really started with backups. Before backup to disk, we used backup to tape. One of the benefits of tape is that backup solutions did not need intelligence around space usage, as tape drives had virtually unlimited capacity – just feed it another tape cartridge. In the 2000s, creating a full weekly (and sometimes full daily) backup and shipping off a carton of tapes on a weekly basis to an offsite location was the norm, thanks to this unlimited capacity paradigm.
However, once backup to disk started becoming popular, that virtually unlimited space disappeared. Eventually, someone had a brilliant idea: what if you had a disk system that scanned the incoming data stream, split it up into chunks, and compared it to what had already been written. Then, if that chunk already existed, the system will just create a pointer to it and move on to the next one.
Voila – the deduplication industry was born!
Admins were initially skeptical, so adoption took a little time. However, it eventually gained acceptance, and as admins got comfortable with it, deduplication became mainstream. Instead of functioning solely as a way to reduce backup disk usage, deduplication started being used for primary data storage.
Here is where I think things get really interesting.
Deduplication as Primary Storage
Going back to our core question up top, part of the reason for deduplication’s acceptance in primary data storage was due to the rigid and inflexible nature of traditional storage platforms in the first place. Many organizations have a storage environment comprised of independent data silos across multiple locations.
Too often, they rely on a heterogeneous mix of platforms that don’t talk to one another (especially in situations that involve M&A), all while trying to maintain performance for users who are just as likely to solve their own problems independent of IT.
Additionally, most of the data needs to be replicated elsewhere for DR purposes, which effectively doubles the storage capacity requirements. Suddenly it’s easy to see why, even after the introduction of deduplication, there are still multiple copies of data floating around within the enterprise. This is completely inefficient.
Think of it in terms of your household tools. If you’re like me, you probably have multiple versions of the same tools spread around your home. Maybe you have a screwdriver or two in the garage, another down in the basement, and a few more in the shed or workshop outside. They may not all be identical sets, but surely there is some overlap.
Now, what if instead of having three or more separate sets, you could have a single tool that is available wherever and whenever you need it, as if by magic? You wouldn’t need 3 or 4 variations of the same tool.
You’d have one tool that’s available everywhere.
The Global File System Alternative
That’s what a global file system can do for you – eliminate the need for multiple copies in the first place. A global file system maintains a single gold copy of data on highly durable media. Now, instead of needing to create independent separate copies of the data where you need it, you just need to provide access to the data where you need it.
In this scenario, files are cached locally at the edge where they are needed, but the central global file system maintains a single, continuously updated gold copy. Thus, you can have a single durable set of data that you can rely on, and have it appear, as if by magic, where and when it is needed.
The global file system does not eliminate deduplication, but it eliminates the friction inherent in traditional storage systems that tends to result in duplication. The difference is that the problem is being solved upstream, negating the need for duplication in the first place, and providing organizations with a much more efficient way to store, protect, and provide access to their data.
The core idea behind deduplication remains a good one, in certain use cases. However, for unstructured file data workloads, a global file system provides some major advantages that can reduce or eliminate the need for deduplication.
Related resources

November 11, 2024 | Ryan Miller
Staying Ahead of Disasters: Improve Business Operations by Modernizing Disaster Recovery
Ryan Miller discusses how changing the type of storage you use can have a significant impact on disaster recovery and improve operations.
Read more
November 06, 2024 | Ryan Miller
Breaking Point: 7 Reasons Traditional Disaster Recovery Falls Short
Nasuni’s Ryan Miller explores the top reasons traditional disaster recovery falls short in large enterprises.
Read more
Webinar
Mastering Modern Backups: Superior Data Storage in the Cloud Era
In episode 2 of Nasuni’s ‘The Power of Cloud Storage’ series, we’re looking at how businesses can overcome multiple file storage challenges by…
Learn more