Enterprise data keeps growing fast. The global data storage market was projected to reach roughly $299 billion in 2026 and is forecast to grow to nearly $985 billion by 2034, according to Fortune Business Insights. Buying more disks alone can't manage growth like that. Storage pooling is what makes large amounts of capacity usable, flexible, and protected against failure.
What is a storage pool?
A storage pool combines the capacity of several physical disks into one shared resource. Instead of managing each disk on its own, IT teams manage the pool as a single unit, then divide it into LUNs for different applications.
Pools typically sit on top of RAID-protected disk groups, so data stays protected even if a drive fails. Administrators can add or remove physical drives from a pool without disrupting the applications that depend on it.
Homogeneous vs. heterogeneous pools
- Homogeneous pool: built from a single drive type: for example, all Serial Attached SCSI (SAS) drives.
- Heterogeneous pool: mixes drive types, such as flash, SAS, and Near Line SAS (NL-SAS), in one pool.
In a heterogeneous pool, frequently accessed (hot) data can sit on fast flash storage, while rarely used (cold) data moves to cheaper SAS or NL-SAS drives. This practice is called storage tiering.
Modern enterprise storage arrays take the heterogeneous pool concept further with automated storage tiering (AST). Rather than manually assigning data to fast or slow drives, the array monitors access frequency for every data block and automatically migrates hot data to the high-performance tier (flash or NVMe) and cold data to the high-capacity tier (SAS or NL-SAS).
Benefits of storage pools
- Higher utilization: Capacity isn't stranded on one server or application.
- Simpler scaling: Admins add drives to the pool instead of building new arrays.
- Mixed drive types: Flash, SAS, and NL-SAS can share one pool.
- Larger scale: A pool can span hundreds of drives.
- Non-disruptive growth: Capacity can be added or reclaimed while the pool stays online.
Storage pools vs. RAID
Storage pools and RAID solve related but different problems. RAID protects a fixed group of drives from failure. A storage pool aggregates capacity: often across multiple RAID groups, so it can be provisioned flexibly as LUNs.
| RAID | Storage pool | |
|---|---|---|
| Drive types | One type per group | Can mix types (heterogeneous pools) |
| Role | The protection layer | Often built on top of one or more RAID groups |
Most storage pools rely on RAID or similar redundancy technologies. For RAID levels and how to choose one, see our guide to RAID configurations.
What is a LUN (logical unit number)?
A LUN is a unique identifier for a defined chunk of storage capacity, following the SCSI addressing standard. When a LUN is presented to a server, the server sees a single physical disk: even though the data may be spread across many drives inside the pool.
Administrators create a LUN by allocating a set amount of pool capacity to it. That capacity can come from one disk or be striped across several. The application using the LUN doesn't need to know which physical disks are involved.
LUN vs. volume
The terms get used interchangeably, but they aren't identical. A LUN is the block-storage identifier a host sees at the SAN layer. A volume is the logical disk an operating system builds on top of a LUN, or on top of several LUNs combined. Most of the time there's a one-to-one relationship, but a single volume can span multiple LUNs.
Pool-based LUN vs. traditional LUN
A traditional LUN is carved directly from one RAID group and is always thick-provisioned. A pool-based LUN draws from a pool that may contain several RAID groups underneath: that structure is what makes thin provisioning, tiering, and larger scale possible.
Types of LUNs
| LUN type | How it works |
|---|---|
| Mirrored LUN | Data is copied to a second physical drive for redundancy. |
| Striped LUN | Data is spread evenly across all drives in the pool, so no single drive is overworked. |
| Striped LUN with parity | Data and parity information are both spread across multiple drives; if one drive fails, its data is rebuilt from parity. |
| Concatenated LUN | Two or more LUNs are joined end-to-end into one larger logical LUN. |
Thin provisioning vs. thick provisioning
Thick provisioning reserves a LUN's full capacity the moment it's created. Thin provisioning allocates space only as data is actually written, and can reclaim that space later if needed.
| Thick provisioning | Thin provisioning | |
|---|---|---|
| Capacity reserved | All of it, upfront | Only what's actually used |
| Unused space | Locked to that LUN | Available to other LUNs in the pool |
| Main risk | Wasted capacity | Pool can run out of space if not monitored |
| Best for | Predictable, steady workloads | Workloads with unpredictable or growing demand |
For example: thick-provision a 100 GB LUN for a workload that only needs 50 GB, and that extra 50 GB stays locked to the LUN: unusable elsewhere. Thin-provision that same LUN instead, and the unused 50 GB stays available to other applications in the pool, until this LUN needs it back.
Thin provisioning is usually the more efficient choice, but it needs closer monitoring. If a thin-provisioned pool runs out of physical capacity, the affected LUN can be forced into a read-only state until space is freed up or added.
LUNs in a SAN
Most pool-based LUNs reach servers over a storage area network (SAN): a dedicated network that connects servers to shared storage systems. Without a SAN, every server would need its own directly attached storage, which makes moving workloads between servers difficult.
A SAN consolidates storage into one system that many servers can share. LUN masking and zoning then control which hosts can see which LUNs. For a full comparison of shared-storage architectures, see our SAN vs. NAS guide.
Access control: LUN masking and zoning
In a SAN, multiple servers connect to a shared storage fabric and can theoretically see every LUN in the environment. Two access control mechanisms prevent one server from accessing another server's storage:
LUN masking is configured on the storage array. It controls which host is permitted to see and access a given LUN. Zoning is configured on the SAN switch. It controls which HBAs can communicate with which storage ports at the fabric level.
Both layers are complementary: zoning limits which devices can communicate across the fabric. LUN masking determines what storage each communicating device is permitted to access.
How storage pools and LUNs work together
Storage pools and LUNs work in sequence, from raw disks to an application-ready volume:
- Physical disks are added to a storage pool.
- The pool groups them, usually with RAID protection underneath.
- An administrator carves out a LUN, sized for a specific application.
- The LUN is presented to a host as a single logical disk.
- The application reads and writes to the LUN without knowing how the underlying disks are arranged.
Storage pool and LUN best practices
- Segment large pools. One giant pool puts every LUN inside it at risk if performance degrades; several smaller pools isolate the impact.
- Watch thin-provisioned pools closely. Set capacity alerts well before the pool fills up — over-subscription is the main risk of thin provisioning.
- Match drives to workload in heterogeneous pools. Keep hot data on flash and cold data on SAS/NL-SAS to get the most from tiering.
- Document the RAID level under each pool. It's easy to forget once the pool abstracts it away, but it still determines your fault tolerance.
- Standardize LUN naming. Consistent names make it far easier to trace a LUN back to its application during troubleshooting.
- Follow vendor-specific guidance. Recommended drive counts, RAID levels, and pool sizes vary by platform.
Monitor storage pools and LUNs with OpManager
Storage problems rarely announce themselves before an application slows down. Monitoring the pool and LUN layer catches issues earlier.
ManageEngine OpManager is a multi-vendor storage monitoring tool that supports a wide range of vendors, including HPE, Dell EMC, IBM, NetApp, Pure Storage, and Hitachi. It gives you centralized visibility into how your capacity is pooled and how each LUN is performing, in a single dashboard.
- Get alerts when capacity, IOPS, latency, or other metrics cross the thresholds you set.
- Use ML-driven forecasting to predict and get alerted when storage capacity is about to approach a critical limit in a certain number of days.
- Monitor RAIDs, tape libraries, and FC switches, along with hardware metrics like fan temperature and voltage.
- View Disks, drives, storage pools, and LUNs in one dashboard, instead of switching between screens.
Interested in learning more?Get a free, personalized demo or request a quote.
FAQs about LUNs, storage pools, and volumes
What does LUN mean?
LUN stands for logical unit number. It's a unique identifier, defined by the SCSI protocol, for a slice of storage capacity that's presented to a server as a single logical disk.
What is a storage pool?
Why use storage pools instead of individual disks?
What's the difference between a LUN and a volume?
What's the difference between a storage pool and RAID?
What happens if a thin-provisioned LUN runs out of space?
Is a LUN the same as a physical disk?
Learn more about storage monitoring
Storage monitoring software
Learn how you can maximize storage uptime and performance with ManageEngine OpManager
Learn more →
By Visakh,
ManageEngine Team