---
# source: src/content/services/en/storage.md
# route:  /en/services/storage/
title: Storage
group: infrastructure
audience: [enterprise, provider]
order: 9
tags: [storage, data-management]
vendors: [hpe]
summary: Block, hyperconverged or scale-out, depending on what runs on it. The choice is settled by recovery time, not by capacity.
photoNeed: "Storage hardware being handled: a disk shelf open, or a drive being seated, hands visible"
stub: false
draft: false
---

## Three shapes, three use cases

**Block storage** for virtual infrastructure, attached over Fibre Channel SAN or iSCSI.
Continuous availability here comes from synchronous replication with automatic failover.

**HCI or dHCI** delivers storage, compute and networking from the same box. The main
advantage is integrated management, and the price is that compute and capacity grow
together.

**Scale-out on software-defined storage** for file services over CIFS and NFS, or for
object storage. Data can sit across several sites, which is how very high availability
figures are reached.

As a backup target, object storage adds the part that matters: **S3 Object Lock** protects
what has been written from being changed, including when production credentials are already
compromised. Spread across several sites, an architecture like that reaches data
availability of up to 14 nines.

## What actually decides it

Usually the recovery time and the number of sites involved, less often the capacity. We
settle those two values at the start, because they determine the shape of the build.

All-flash with NVMe brings the latency down. It does nothing for availability; replication
and failover do that.

Hybrid systems with flash and disk remain the right choice where capacity matters more than
latency.