Tier Your Unstructured Data to Lower Storage Costs and Cyber Risk

This guest blog was contributed by Kumar Goswami is CEO and co-founder of Komprise

Unstructured data has quietly become a whale-like burden on IT infrastructure, storage, and security resources. Large enterprises commonly house 20, 30 or 40 petabytes (PB) of unstructured data, which is primarily generated by applications and users and ends up locked in file data storage systems that consume a disproportionate share of IT budgets. This file data quagmire is also risky business, particularly in the face of growing ransomware threats.

For IT leaders, the rapid growth of file data is no longer a background issue—it’s front and center. Cost optimization and risk management have become top priorities in enterprise data strategies. File-level unstructured data tiering (a form of online archiving) has emerged as a smart and effective solution.

Why File Data Now Commands CIO Attention

According to the 2024 Komprise State of Unstructured Data Management survey, more than half of CIOs—56%—identify cost optimization as their top data management priority. This makes sense when considering the nature of file data. Unlike structured databases of rows and columns, file data often consists of documents, images, videos, and logs that can be retained for decades, typically without any clear data lifecycle policies. In addition, files can take up large swaths of prime storage space.

This long-term accumulation leads to multiple redundant copies: a primary version, a backup, and a disaster recovery (DR) version. This of course can triple storage requirements and associated costs. Ironically, much of this data is rarely accessed or “cold,” yet may be sitting on expensive storage devices in the data center or in high-performance file storage in the cloud.

At the same time, business leaders are realizing that file data is an untapped asset for analytics and artificial intelligence. When managed intelligently, this data can improve customer relationships, support product innovation and inform strategic decisions. But doing so affordably and securely requires a shift in how file data is stored and protected.

The Ransomware Threat to File Data

File data is particularly vulnerable to ransomware attacks because it is widely accessed across users, groups, and systems. This broad exposure means that even one compromised user account can lead to a widespread infection. Since file systems are often interconnected, ransomware can silently spread through the network before being detected.

Given the complexity and distributed nature of file data, it’s no surprise that it represents one of the largest surfaces for potential ransomware damage. Ignoring this exposure is no longer an option. A comprehensive ransomware strategy must include file data protection—and that’s where data tiering becomes crucial.

What Is File-Level Data Tiering?

File-level data tiering is a method of reducing the cost and risk of storing cold file data while ensuring a non-disruptive experience for data access and ongoing mobility. An unstructured data management system first scans files across storage and then identifies files that are no longer active. The timeframe for labeling or “tagging” data as ”cold" could be three months for medical images and closed legal cases or one year for user documents.

The data management system then moves the tagged cold data from expensive primary storage (a.k.a. a network attached storage or NAS system) to an economical secondary location, such as cloud object storage, which has higher latency but a fraction of the cost per terabyte (TB). Unlike block-level tiering (which happens behind the scenes within storage systems), file-level tiering operates on entire files and should deliver a transparent experience for end users and applications.

Tiered files still appear in the same folder structure and can be opened like normal, but the actual data is stored elsewhere. A user may never know the difference. This transparency means there’s no need to change applications or user workflows.

The Financial and Security Benefits

Offloading cold files from high-performance storage to a more cost-effective archive platform offers dramatic savings. By reducing the active storage, backup, and DR footprint, organizations can cut total storage-related costs by more than 70% annually.

Here’s an example based on a 1-PB environment:

Cold Data Identified: 80% of total volume (819TB)
Annual Storage Costs (Traditional Setup): $2.59 million USD
Annual Costs with Tiering in Place: $770,000 USD
Annual Savings: $1.82 million USD, or 70%

Beyond cost savings, tiering also shrinks the ransomware attack surface. Since cold files are no longer actively stored in the primary environment, they are effectively removed from the reach of most threats. This shift can reduce the volume of vulnerable data by up to 80%, significantly lowering both the probability and impact of an attack. Additionally, immutable storage options can be used for cold data, preventing modifications and deletions and further enhancing ransomware resilience.

There is the added benefit that one can deploy a more extensive (and likely more expensive) ransomware solution since it’s only needed on 20% of the data.

Tiering Without Disruption Boosts Productivity & Departmental Buy-In

Unlike traditional data migrations or storage refreshes, file data tiering operates quietly in the background. File access remains unchanged for users and applications, ensuring operational continuity while silently improving security and reducing costs.

Ideally, an unstructured data management solution can suit the organization’s needs, allowing IT to set policies for identifying which files to tier based on age, usage patterns, file types, or ownership. However, there is even more power and potential cost savings when authorized departmental users review their own files and tag data sets which they deem ready for tiering.

This allows IT leaders to optimize tiering strategies in a collaborative fashion, maximizing cost savings to the larger organization--and departments specifically--if they are on a chargeback plan. As well, a non-disruptive tiering technique that allows for direct, native access at the destination means that researchers, data scientists and other data stakeholders can rest easy; the data that they designate for archives will be readily available when and if they need it down the line for an AI or big data analytics projects.

In a time when CIOs and other IT leaders are under pressure to be more efficient, avoid costs wherever possible while reducing security risks and preparing data for AI, there is a lot at stake. The practical idea of file-level data tiering offers a high-impact solution that can preserve the data estate for competitive advantage.

To review, this delivers:

Cost Optimization: By removing inactive data from expensive storage environments and backup workflows and by reducing the amount of data to be protected from ransomware, you can save 70-80% on storage and backups annually.
Cost Avoidance: By reducing the capacity of the data stored on expensive primary storage, you avoid buying additional primary storage. In today’s market with rising prices due to tariffs, this is a nice additional insurance to have.
Risk Reduction: By limiting the attack surface and exposure of sensitive or stale files, you have an expanded ransomware defense.
Future-Readiness: By enabling secure, low-cost retention of data, you are creating valuable unstructured data stores for AI and analytics initiatives.

Final Thoughts

File data has quietly become one of the most costly and risky assets in the enterprise. But with smart, automated strategies like file-level data tiering, CIOs can take back control—dramatically reducing costs, boosting security, and preparing their organizations for AI.