Glossary
Storage management
Storage management is the ongoing work of running an organization's storage: provisioning it, protecting it, monitoring it, governing access and data lifecycle, and planning capacity. Almost every one of those tasks depends on a capacity figure, and the same hardware can honestly be described by three different ones.
Six duties that make up the work
- Provisioning. Creating volumes, shares and buckets and assigning capacity to the teams that ask for it.
- Protection. Choosing how data survives drive, server and site failures, and how it is backed up.
- Monitoring. Tracking health, performance, capacity and the rate at which space is consumed.
- Lifecycle. Moving and expiring data as it ages, covered under storage lifecycle management.
- Access control. Deciding who can read, change or delete which data.
- Capacity planning. Forecasting growth early enough that new hardware arrives before space runs out.
Raw, usable and effective capacity
| Figure | What it counts | What reduces it |
|---|---|---|
| Raw | The total of every drive | Nothing; it is the starting point |
| Usable | What remains after protection takes its share | Copies or parity, spare space, metadata |
| Effective | What can be stored after data reduction | Data that does not compress or deduplicate |
One petabyte of raw drives protected by three full copies gives about 333 TB usable. The same petabyte protected by erasure coding with eight data pieces and four parity pieces gives about 667 TB, before spare capacity and metadata. Same hardware, same price, twice the stored data. A figure quoted with no label is usually raw.
Where do wrong figures cause damage?
Alerts, forecasts, chargeback and purchasing all start from a capacity number. A forecast built on raw capacity runs out early. An effective figure built on a reduction ratio from a datasheet rather than the organization's own data overstates what fits, a point taken further under storage compression. In a thin-provisioned pool, allocated capacity says nothing about physical space left.
How the workload scales
Each separate system has its own console, alerts and spare space, so storage sprawl multiplies the work. Shared pools reduce the number of places to watch, and storage automation takes over the repetitive steps, leaving people with the decisions about protection, retention and growth.
Frequently asked questions
What is the difference between raw and usable capacity?
Raw is the sum of all drive capacity. Usable is what remains once replicas or parity, spare space and metadata are taken out, so it is always lower.
How far ahead does storage capacity planning look?
At least as far as it takes to approve, order, deliver and install hardware. That lead time, not the current free space, decides when a capacity alert has to fire.