Storage costs money. Not just in the obvious sense of buying drives or paying cloud bills – but in backup time, disaster recovery complexity, and the IT hours spent managing a file server that’s perpetually two weeks away from running out of space. For most organisations, the root cause of all of this is the same: everything lives in the same place, treated with the same priority, regardless of how often it’s actually used.
A tiered storage strategy changes that. By deliberately separating data based on how frequently it’s accessed, you stop paying premium prices to store files nobody has opened in three years. Done well, it’s one of the highest-return infrastructure decisions an IT team can make – and it’s far more accessible than it used to be.
This article walks through how to think about tiered storage, how to build a practical strategy, and what to watch out for along the way.
The Core Idea: Not All Data Is Equal
Most organisations, if they’re honest, have a file server that looks something like this: a shared drive full of folders, accumulated over years, where roughly 20% of the files are accessed regularly and the other 80% haven’t been touched in months or years. Yet everything sits on the same expensive, fast, well-backed-up primary storage.
Tiered storage is the recognition that different data has different value – not in terms of importance, but in terms of access frequency. A contract signed last week needs to be fast and immediately available. A contract signed in 2019 needs to be findable and accessible, but probably doesn’t need to live on your highest-performance storage.
The basic principle is to match storage cost to access frequency:
- Frequently accessed data → fast, expensive primary storage
- Infrequently accessed data → slower, cheaper secondary storage
- Rarely or never accessed data → cold storage, archival, or scheduled deletion
This isn’t a new concept. Hierarchical Storage Management (HSM) systems have been doing this for decades in enterprise environments. What’s changed is that the tools to implement it have become far more affordable and far less complex.
Step 1: Understand What You’re Actually Storing
Before you can build a tiered strategy, you need visibility into your current storage. This means running an audit – either with dedicated tooling or with a file archiving solution that can analyse your file system before doing anything else.
Key questions your audit should answer:
- What is the age distribution of your files? How much of your data hasn’t been accessed in over a year? Over three years?
- What are your largest files, and how frequently are they accessed? Large, static files (old video exports, archived project folders, legacy databases) are often ideal candidates for early archiving.
- Are there file types that are never actively used? Old installer packages, duplicate exports, and superseded document versions often accumulate silently.
- Which departments or users are generating the most data? This can inform policy decisions about retention and archiving thresholds.
Most organisations are surprised by what this reveals. It’s common to find that 60–70% of stored data qualifies as cold – files that haven’t been touched in over a year. That’s a significant proportion of your storage spend that could be moved somewhere cheaper without any meaningful impact on daily operations.
Step 2: Define Your Storage Tiers
Once you understand your data landscape, you can design your tiers. There’s no single right answer here – the right structure depends on your infrastructure, budget, and compliance requirements. A typical three-tier model looks like this:
Tier 1 – Live Storage Your primary file server or SAN. Fast, resilient, regularly backed up. This is where active files live – anything accessed in the last 30, 60, or 90 days, depending on your policy.
Tier 2 – Near-Line Archive A secondary storage destination that’s still readily accessible but cheaper than Tier 1. This could be a NAS device, a lower-cost SAN volume, or cloud object storage. Files moved here are still accessible to users – typically via symbolic links or stubs left behind in Tier 1 – but they’re no longer taking up space on your expensive primary storage.
Tier 3 – Cold Storage or Graveyard For data that needs to be retained for compliance reasons but is extremely unlikely to be accessed. This might be cloud cold storage (Amazon Glacier, Azure Archive), tape, or a dedicated archive volume. Some organisations add a deletion policy at this stage – files that have sat in cold storage for a defined period are reviewed and purged.
This three-tier model mirrors what solutions like ArchiverFS from MLtek implement natively, with Live, Archive, and Graveyard stages built into the product’s architecture. The principle, however, applies regardless of which tool you use to implement it.
Step 3: Set Your Archiving Policies
A tiered storage strategy without clear policies is just a complicated file copy job. The policies are what make it automatic, consistent, and defensible – particularly from a compliance perspective.
Policies typically define:
- Age thresholds – files not accessed in X days move from Tier 1 to Tier 2; files not accessed in Y days move from Tier 2 to Tier 3
- Size rules – large files above a certain threshold may be candidates for earlier archiving regardless of age
- File type exclusions – some file types (active database files, system files, recently modified documents) should be excluded from automatic archiving
- Department or folder exceptions – some areas of the file system may need different treatment based on business requirements
The key is to set policies that reflect actual business needs rather than arbitrary numbers. If your legal team regularly references documents from three years ago, a 12-month archiving threshold for that folder will cause frustration. Talk to stakeholders before finalising policies.
Step 4: Choose Your Archiving Tool
The tool you use to implement tiered storage will significantly affect how smooth the process is – both for IT and for end users. There are several categories of solution to consider:
Manual processes – scripts and scheduled tasks that move files based on age. Low cost, high maintenance, and prone to breaking silently. Not recommended beyond very small deployments.
Built-in Windows tools – Windows Server has some file management capabilities, but they’re limited and don’t handle user transparency (stub files) well.
Dedicated file archiving software – purpose-built tools that handle policy management, stub creation, metadata preservation, and reporting. This is the practical choice for most organisations.
Full HSM platforms – enterprise-grade systems with sophisticated tiering capabilities, typically accompanied by enterprise-grade price tags. Appropriate for very large or complex deployments, but overkill for most.
When evaluating dedicated archiving tools, the architecture questions matter as much as the feature list. Look for solutions that preserve NTFS permissions and timestamps, create stubs using standard file system constructs rather than proprietary formats, and support your target storage destinations – whether that’s on-premises NAS, cloud storage, or both.
Step 5: Plan for User Transparency
A tiered storage strategy only works in practice if users can still access their files without friction. This is non-negotiable. If people start encountering errors, missing files, or slow recalls when trying to open archived content, the strategy will lose organisational trust quickly.
The mechanism that handles this is typically a stub or symbolic link left behind when a file is moved. When a user clicks on it, they get the file – ideally without any perceptible delay.
ArchiverFS handles this with direct-access links, meaning IO goes straight from the user’s machine to the archive storage without any intermediary recall process. Not all tools work this way – some require a full file recall to primary storage before the user can open it, which introduces noticeable latency for large files. It’s worth testing this specifically during any product evaluation.
Step 6: Integrate With Your Backup Strategy
One of the less obvious benefits of tiered storage is what it does to your backup window. If 70% of your data is cold and living in Tier 2 or Tier 3 storage, you can back up Tier 1 far more frequently and efficiently – because it’s a fraction of the size.
Cold archive data doesn’t change. You can back it up weekly or monthly rather than nightly, which reduces both backup duration and storage costs. Some organisations take this further and treat their Tier 2 archive as a secondary resilience layer in its own right – particularly if it’s replicated cloud storage.
Coordinate with whoever manages your backups before finalising your tiering strategy. The interaction between archiving stubs and backup software can be nuanced, and it’s worth validating behaviour in a test environment first.
Common Pitfalls to Avoid
Even well-planned tiered storage strategies can run into problems. The most common:
- Setting policies too aggressively – archiving files that people need regularly erodes trust fast. Start conservative and tighten thresholds over time.
- Ignoring compliance requirements – some industries have mandatory retention periods. Make sure your Tier 3 deletion policies account for these.
- Underestimating Tier 2 storage needs – the whole point is to move data off Tier 1, which means Tier 2 needs to be large enough to absorb it. Plan for growth.
- Not testing user access before rollout – stub behaviour can vary by application and OS version. Test broadly before going live.
- Choosing a tool with proprietary storage formats – if the software traps your archived files in a database or custom format, you’re trading one problem for another. Ensure archived files remain accessible as standard files on standard storage.
The Payoff
A well-implemented tiered storage strategy typically delivers several immediate and ongoing benefits: lower primary storage costs, faster backup windows, reduced disaster recovery complexity, and a file server that doesn’t require emergency capacity upgrades every eighteen months.
Perhaps more importantly, it gives your organisation a sustainable model for managing data growth – one that scales without requiring proportional increases in storage spend. As data volumes continue to grow, that scalability becomes increasingly valuable.
The technology to do this properly is no longer the preserve of large enterprises with specialist teams and substantial budgets. With the right tooling and a clear policy framework, organisations of almost any size can implement genuine tiered storage – and start seeing the benefits quickly.
The hardest part, usually, is getting started. Run the audit. See what you’re actually storing. The rest tends to follow naturally.
Apart from that, if you want to know more about Why Infrastructure Location Still Matters in the Cloud Era then visit our Tech category.
