Komprise Hybrid Tiering:
Unlock the Potential of Your Unstructured Data

5 considerations when picking a storage tiering solution

This paper introduces hybrid tiering, which tiers unstructured data across your entire hybrid storage infrastructure including on-premises and cloud, file and object storage.

Hybrid tiering is an independent, storage-agnostic intelligent tiering solution which can:

1. Reduce your ransomware attack surface;
2. Avoid rehydration when refreshing your storage;
3. Provide flexible data management polices while maintaining transparent access;
4. Ensure innovation and choice with cloud native data access;
5. Ensure your unstructured data is ready for AI.

Download White Paper Now

Read this paper to understand the benefits of Komprise hybrid tiering vs. storage-based tiering.

This paper reviews these five considerations when evaluating data tiering options and presents a cost model that outlines the benefits of hybrid tiering as both a complement and an alternative to built-in tiering capabilities from data storage vendors.

Intelligent Tiering FAQs

What is hybrid data tiering and how is it different from built-in storage vendor tiering?

Hybrid data tiering is a storage-agnostic approach to moving unstructured data across an enterprise’s entire hybrid storage infrastructure — on-premises NAS, cloud object storage, and secondary tiers, based on policy-driven rules, without disrupting user or application access. It differs fundamentally from built-in storage vendor tiering in five ways:

• Storage-agnostic vs. single-vendor — hybrid tiering works across any NAS vendor and any cloud provider; storage-vendor tiering is designed to optimize within that vendor’s own system only
• File-level vs. block-level — hybrid tiering moves entire files with full metadata, permissions, and attributes intact; vendor tiering moves blocks that are meaningless without the originating file system
• No lock-in — built-in storage-based tiering solutions lock data in a proprietary block format, resulting in 75% higher cloud egress costs and lock-in; hybrid tiering writes data in open standards with no vendor dependency
• Cloud-native access — tiered data remains natively accessible as files or objects from the cloud destination without routing through the source NAS; vendor-tiered blocks require the proprietary storage OS to be running
• No rehydration penalty — hybrid tiering avoids rehydration by maintaining file fidelity and portability across storage platforms; vendor tiering triggers costly recalls during antivirus scans, backups, and hardware refreshes

How can hybrid data tiering help IT teams offset the surge in NAND flash and SSD storage prices?

NAND flash and DRAM prices are projected to rise up to 130% by end of 2026 due to ongoing memory supply shortages, compressing storage budgets at exactly the moment AI and data workloads are demanding more capacity. Hybrid data tiering directly addresses this through Komprise Flash Stretch:

• Free 70%+ of flash capacity — over 70% of data on expensive NAS has not been accessed in months yet consumes the same backup resources as active data; tiering this cold data off flash frees immediate headroom without a hardware purchase
• No new procurement required — Komprise Flash Stretch identifies cold data on expensive primary storage and moves it transparently to lower-cost object storage or cloud, stretching existing flash capacity rather than replacing it
• Eliminate rehydration costs — unlike storage-vendor tiering that triggers unexpected recalls during antivirus scans or defragmentation, driving egress fees and forcing more storage purchases, Komprise Transparent Move Technology eliminates rehydration entirely
• Fast ROI — Pfizer reduced storage, backup, and DR costs by 70–75% with full payback within 90 days
• Future-proof against further price volatility — since data often outlives storage hardware, hybrid tiering prevents costly rehydration during vendor or hardware refreshes by keeping files portable and storage-agnostic

How does hybrid data tiering reduce ransomware risk alongside cutting storage costs?

Hybrid data tiering, also known as Intelligent Tiering, delivers ransomware defense as a direct byproduct of cost optimization — not as a separate security tool. The mechanism is straightforward:

• Shrinks the attack surface — by offloading cold data from the active footprint and leveraging object-locked destinations, hybrid tiering reduces the ransomware attack surface by up to 70%
• Immutable cloud destinations — Komprise tiers cold data to immutable object storage such as Azure Blob with versioning or Amazon S3 Object Lock; even if cold files are targeted, they are saved as new versions — enabling clean recovery from the prior version
• Reduces backup blast radius — cold data tiered off primary NAS is also removed from the backup footprint, cutting backup windows and DR costs while simultaneously reducing the volume of data exposed to ransomware
• No proprietary dependency at recovery — because Komprise writes data in native file and object formats, recovery does not require the source storage vendor’s software to be intact — critical when a ransomware event may have compromised the storage system itself
• Dual benefit in one motion — offloading cold files to immutable Azure Blob eliminates 80% of storage, DR, and backup costs while also providing a potential recovery path if cold files are infected

Learn more

Why does cloud-native data access matter for AI, and how does hybrid tiering enable it?

Cloud-native data access, the ability to read, process, and use tiered data directly from its cloud destination in native file and object formats, is increasingly a prerequisite for enterprise AI initiatives. Most enterprise unstructured data starts on NAS and must be AI-ready at cloud scale to be useful. Hybrid tiering enables this in four ways:

• No proprietary format lock-in — with block-based tiering, the only way to access data in the cloud is to run the proprietary storage file system in the cloud, which adds cost; with native file-level tiering, data is directly accessible from the target storage for AI services without any intermediary
• Direct AI pipeline integration — data tiered to cloud object storage (AWS S3, Azure Blob, Google Cloud Storage) is immediately accessible to cloud-native AI training, inferencing, RAG pipelines, and analytics services without a secondary migration or ETL step
• Global Metadatabase foundation — Komprise indexes all file metadata at the point of tiering, building the Global Metadatabase with location, access pattern, file type, and classification data that Smart Data Workflows use to identify and curate AI-ready datasets across the full hybrid estate
• Filters AI pipeline noise — intelligent tiering identifies 60–80% of unstructured data that no longer needs expensive, high-performance file storage; removing this cold, duplicate, and irrelevant data from primary storage delivers a cleaner, higher-signal dataset to AI pipelines, directly improving model accuracy and reducing compute costs
• Agentic AI compatibility — a well-tiered, indexed data estate gives AI agents cross-silo visibility to locate and retrieve the right data at runtime without manual intervention

What are key considerations when choosing a storage tiering solution, and how does Komprise address each one?

The Komprise Hybrid Tiering white paper identifies five considerations when evaluating data tiering options, each one a failure mode of proprietary vendor-based tiering that hybrid tiering solves:

• Reduce your ransomware attack surface — Komprise tiers cold data to immutable object storage by policy, shrinking the active data footprint exposed to ransomware by up to 70% while simultaneously cutting storage and backup costs
• Avoid rehydration during storage refreshes — when hardware reaches end-of-life or a vendor contract ends, proprietary block-based tiering requires buying more of the same vendor’s storage to rehydrate data before migrating; Komprise Transparent Move Technology keeps files portable and format-agnostic, eliminating rehydration entirely
• Provide flexible management policies with transparent access — Komprise supports custom policies across file type, age, owner, directory, project code, and last access date across any combination of NAS and cloud; users and applications always access files from their original location with no workflow disruption
• Ensure cloud-native data access and choice — adopting a flexible data tiering solution that delivers maximum ROI is paramount — not only because storage budgets are stretched too thin but because intelligent data tiering supports feeding AI data pipelines; Komprise writes data in open standards to any cloud, ensuring no future lock-in
• Ensure unstructured data is ready for AI — in a 1PB scenario, Komprise hybrid tiering cut annual costs by 75% compared to storage-based tiering, while simultaneously indexing the full data estate in the Global Metadatabase — creating the governed, searchable metadata foundation that AI pipelines require to identify and ingest the right data at petabyte scale