Back to blog

What is object storage and why it beats a NAS for most use cases

If you have ever managed files for a team, backed up a photography library or stored media for a project, you have probably used a NAS (Network Attached Storage) or at least considered buying one. NAS devices have been the default answer to “where do we put our files” for decades. They sit in a closet, connect to the local network and give everyone a shared drive.

But over the last few years, a different approach has been quietly replacing NAS setups and traditional file servers across businesses of every size. It is called object storage, and services like Amazon S3, Cloudflare R2, DigitalOcean Spaces and Wasabi have made it accessible to anyone with an internet connection. Here is what object storage actually is, how it compares to the alternatives and why it is winning.

What is object storage?

Object storage is a way of storing data as individual objects rather than organizing them into a traditional file-and-folder hierarchy. Each object is a bundle of three things: the data itself (a photo, a video, a document), a unique identifier (its key) and metadata (information about the object, like its size, content type and creation date).

Unlike the file system on your computer or NAS, object storage uses a flat namespace. There are no real directories. When you see a path like photos/2026/march/sunset.jpg, that entire string is just the object’s key. The slashes are part of the name, not actual folder boundaries. This is a subtle distinction, but it is what allows object storage to scale without the overhead of maintaining a directory tree.

You access objects through HTTP-based APIs, not through a mounted drive. The most widely adopted API is the one Amazon created for S3, and it has become a de facto standard. Cloudflare R2, DigitalOcean Spaces, Wasabi and dozens of other providers all implement the S3 API, which means tools and code built for S3 work across providers with minimal changes.

How it differs from NAS and block storage

To understand why object storage matters, it helps to see how the three main storage types compare.

File storage (NAS). A NAS presents storage as a hierarchical file system over the network, usually using protocols like SMB or NFS. Your computer mounts the NAS as a network drive, and you interact with it exactly like a local disk — creating folders, moving files, renaming things. The NAS manages the directory structure, file permissions and locks. This works well for small teams and local access, but it becomes harder to scale and more expensive to maintain as data grows.

Block storage (SAN/EBS). Block storage divides data into fixed-size blocks and stores them without any inherent structure. It is the foundation for databases, virtual machine disks and any application that needs raw, high-performance I/O. AWS EBS (Elastic Block Store) is a common example. Block storage is fast but expensive, and it is typically attached to a single server. You would not use it for general-purpose file storage.

Object storage (S3, R2, Spaces, Wasabi). Object storage treats every piece of data as an independent object in a flat namespace. It is accessed over HTTP, scales to petabytes without any configuration changes and charges based on how much you store rather than requiring you to provision capacity in advance.

Comparison table

Object storageFile storage (NAS)Block storage (SAN/EBS)
StructureFlat namespace, key-valueHierarchical (folders and files)Raw blocks, no structure
Access methodHTTP / S3 APISMB, NFS, AFPiSCSI, Fibre Channel, NVMe
ScalabilityPetabytes+, virtually unlimitedLimited by hardware capacityLimited by provisioned volume
Cost modelPay per GB stored + requestsUpfront hardware + maintenancePay per provisioned GB
RedundancyBuilt-in (multi-AZ, 11 nines on S3)Depends on RAID configDepends on provider/config
MaintenanceNone (fully managed)Hardware replacements, firmware, backupsProvider-managed or self-managed
Best forMedia, backups, archives, web assetsLocal file sharing, POSIX appsDatabases, VMs, high-IOPS apps

Why object storage is winning

Several trends have converged to make object storage the default choice for the majority of file storage needs.

Scales without managing disks

A NAS has a fixed number of drive bays. When you run out of space, you either replace drives with larger ones, add an expansion unit or buy a new NAS entirely. With object storage, you upload another terabyte and your monthly bill increases by a few dollars. There is no capacity planning, no downtime for upgrades and no limit you need to think about in advance.

Built-in redundancy

Amazon S3 is designed for 99.999999999% (eleven nines) durability. That means if you store ten million objects, you can statistically expect to lose one every ten thousand years. This level of durability is achieved by automatically replicating data across multiple facilities. Achieving anything close to that with a NAS would require a complex multi-site replication setup that most teams simply do not have the resources to build.

No hardware to maintain

Hard drives fail. RAID arrays degrade. Firmware needs updating. NAS units reach end of life. With object storage, all of that is the provider’s problem. You never think about disk health, power supply redundancy or replacing a failed drive at 2 AM.

Pay for what you use

A NAS requires an upfront investment regardless of how much data you actually store. You buy a four-bay unit, populate it with four drives and pay the full cost on day one even if you only use 10% of the capacity for the first year. Object storage charges per gigabyte per month with no minimum commitment on most providers.

The S3 API is a universal standard

Because Amazon’s S3 API has been adopted by virtually every alternative provider, you are not locked into a single vendor. If you start with AWS and later decide Cloudflare R2 offers better pricing for your use case, you can migrate without rewriting your tools. For a deeper comparison between the major providers, see our S3 vs R2 vs Spaces vs Wasabi breakdown.

When a NAS still makes sense

Object storage is not the right answer for every situation. There are cases where a local NAS remains the better choice.

Low-latency local access. If your workflow depends on accessing large files with minimal latency — for example, video editors working with raw footage on a local network — a NAS connected over 10 Gigabit Ethernet will outperform any cloud storage. Object storage access goes through the internet, and even fast connections introduce latency that matters for real-time editing.

POSIX file system requirements. Some applications expect a traditional file system: they need to lock files, append to them in place, follow symbolic links or traverse directory trees. Object storage does not support these operations. If your software requires a mounted volume with full file system semantics, a NAS or block storage is what you need.

Small office with existing infrastructure. If you already have a NAS that handles your current storage needs and you do not expect significant growth, there may be no compelling reason to migrate. The cost of the hardware is already spent, and the setup works. Migration makes sense when the NAS is nearing capacity, the hardware is aging or you need access from outside the local network.

The missing piece: a user-friendly interface

There is one significant gap in the object storage experience. Providers like S3, R2, Spaces and Wasabi give you powerful, scalable storage, but they do not give you a convenient way to work with your files day to day. The AWS console is designed for developers. The Cloudflare dashboard is functional but minimal. There is no native equivalent of opening a folder on your desktop, browsing thumbnails and dragging files around.

Nubbo dashboard showing the file browser in list view with files, sizes and dates

This is the problem Nubbo solves. Nubbo is a web-based management layer that connects to your own S3-compatible storage and provides a familiar file browser experience on top of it. You get features like file sharing with password protection and expiration dates, photo and video galleries for visual projects, file requests so clients can upload directly to your bucket, a built-in media player for audio and video, and labels and keyboard shortcuts for efficient organization.

The key architectural detail: your files never pass through Nubbo’s servers. All transfers happen directly between your browser and your storage provider using presigned URLs. Your credentials are encrypted with AES-256-GCM, and your account is protected with two-factor authentication. You can read more about this on the security page.

Start managing your object storage visually

Object storage is the most practical, cost-effective and resilient way to store files for the vast majority of use cases. The only thing it has been missing is a good interface. If you already use S3, R2, Spaces or Wasabi — or you are thinking about moving away from a NAS — Nubbo gives you the visual layer that makes object storage feel as natural as a local drive.

Create your free account and connect your first bucket in under two minutes.