HPE Alletra Storage
HPE Alletra: all-flash storage arrays for databases, and containers. Latency under 100 microseconds.

Key Features
- All-flash array (SSD/NVMe drives only)
- Latency <100μs for fastest applications
- InfoSight predicts drive failures 24h in advance
- Capacity up to 75 PB after compression and deduplication
- Connectivity via FC, iSCSI, or NVMe-oF
Table of Contents
What is HPE Alletra?
HPE Alletra is a storage array - a device for storing data for servers. You connect servers to it (via FC or Ethernet network), and they see disks where they keep data - virtual machines, SQL/Oracle databases, user files.
Why “all-flash”? Because inside there are only SSD/NVMe drives - no traditional spinning HDDs. Thanks to this, Alletra is very fast - latency (response time) below 100 microseconds. For comparison, a traditional array with HDDs has latency of 5-10 milliseconds - 50-100 times slower.
What is it used for?
Alletra is storage for applications requiring fast data access:
- Hyper-V - storing virtual machines
- Databases - SQL Server, Oracle, PostgreSQL, SAP HANA
- Containers - persistent storage for Kubernetes
- VDI - virtual desktops (Citrix)
Typical scenario: You have 50 virtualization servers. Instead of putting disks in each server, you connect them to one Alletra array. All VMs keep data on Alletra. You gain: centralized backup, replication to a second location, easy VM migration between servers.
What is the difference between models?
HPE offers several Alletra variants - from small businesses to the largest banks:
| Model | For whom | Capacity | Latency | Approximate price |
|---|---|---|---|---|
| Alletra 5000 | Small/medium businesses | Up to 1.5 PB | <250μs | From ~$35K |
| Alletra 6000 | Medium businesses | Up to 4 PB | <150μs | From ~$70K |
| Alletra 9000 | Enterprise, banks | Up to 75 PB | <100μs | From ~$230K |
| Alletra MP | Unified (block+file+S3) | Up to 35 PB | <150μs | From ~$115K |
“Effective” capacity = after compression and deduplication. If you have 100 TB of raw data, Alletra can compress it to ~30 TB of physical disks.
What is HPE InfoSight?
InfoSight is a monitoring system built into every Alletra:
- Collects data from the array - temperature, disk load, errors, latency
- Sends to HPE cloud - where AI analyzes data from 150,000 other arrays
- Predicts problems - e.g., “disk #5 shows failure pattern, replace within 48h”
- Opens tickets - HPE automatically opens a support ticket and sends a replacement disk
Effect: 86% of problems are resolved before you notice something is wrong.
How is it connected?
Alletra connects to servers via storage network:
- Fibre Channel (FC) - dedicated fiber optics 16/32 Gb/s - lowest latency
- iSCSI - via regular Ethernet network 10/25 GbE - simpler, cheaper
- NVMe-oF - newest protocol via Ethernet or FC - fastest
Example configuration:
[Server 1] ----\
[Server 2] -----+--[FC Switch]--[Alletra 6000]
[Server 3] ----/
Replication and backup
Alletra has built-in data protection features:
- Snapshots - data snapshots every minute/hour, no performance impact
- Synchronous replication - data written simultaneously to 2 arrays (for banks)
- Asynchronous replication - copy to second location with minute delay (for DR)
- Veeam integration - VM backup directly from array (faster than via network)
Specifications (Alletra 6000)
| Drives | NVMe SSD, from 1.92 TB to 30.72 TB per disk |
| Controllers | 2 active-active (one fails = continuous operation) |
| Raw capacity | Up to 1.3 PB |
| Effective capacity | Up to 4 PB (at 3:1 compression) |
| IOPS | Up to 2.4 million operations/second |
| Latency | <150μs |
| Protocols | FC 32Gb, iSCSI 25GbE, NVMe-oF |
For whom?
Alletra makes sense when:
- You have applications requiring low latency (databases, VDI)
- You need centralized storage for multiple servers
- You want replication to a second location
- You care about predictable performance (99.9999% SLA)
Alletra does NOT make sense when:
- You only need archive storage (better: HPE StoreOnce or tapes)
- You have a small budget and NAS is enough (better: Synology/QNAP)
- Cloud storage is sufficient (AWS S3, Azure Blob)
FAQ
How does Alletra differ from NAS? NAS shares folders (SMB/NFS). Alletra shares “disks” (block storage via iSCSI/FC). Alletra is faster and better for VMs/databases. NAS is simpler for user files.
How does Alletra differ from Nimble? Nimble is HPE’s older line (hybrid SSD+HDD). Alletra is newer, purely all-flash. For new deployments - Alletra. Nimble still sold for existing customers.
How much does it cost? From ~$35K for the smallest Alletra 5000 to several million for Alletra 9000. Specific price depends on capacity and configuration.
Can I start small and expand? Yes. Drives are added during operation (hot-plug). Some models also allow adding controllers.
What is the warranty? 3 years standard, up to 5 years with extension. HPE guarantees 99.9999% availability (max 32 seconds downtime/year).
Do I need an internet connection? For InfoSight - yes. Without internet, the array works, but you lose predictive analytics.
What does support look like? HPE has 24/7 support. InfoSight often detects a problem and opens a ticket before you call.
Can I use my own disks? No. Alletra requires HPE disks - this guarantees performance and InfoSight operation.
How long does deployment take? Typically 1-2 days for physical installation + configuration. Data migration - depends on amount.
Is there a “trial” version? HPE offers PoC (Proof of Concept) - you can test Alletra in your environment before purchase.
Inquire about HPE Alletra Storage
Contact your product specialist and get a custom quote.

Related Services
Our services supporting the implementation and management of this solution
Comprehensive Storage Services
IT Infrastructure
Build storage that won't fail at critical moment. Design, implementation, support from storage specialists.
Storage System Implementation
IT Infrastructure
Fast storage with enterprise redundancy. Performance matched to workloads, not budget.
Comprehensive Network Infrastructure Implementation
IT Infrastructure
Build a network that doesn't fail. From design through implementation to 24/7 support.
Server Infrastructure Implementation and Optimization
IT Infrastructure
Match servers to actual needs. Save 30% by avoiding over-provisioning.
From Our Knowledge Base
Articles related to this solution
CVE-2026-42809: Privilege escalation in Apache Polaris via vended storage credentials
Security Alert - CVE-2026-42809 (Apache Software Foundation Apache Polaris). CVSS: 9.9 (critical).
CVE-2026-5976: OS command injection via setStorageCfg() in Totolink A7100RU CGI
Security Alert - CVE-2026-5976 (Totolink A7100RU). CVSS: 9.8 (critical).
Network Design — How to Build a Secure Network Infrastructure
Network design is the process of planning and structuring a computer network to meet performance, scalability, and security requirements. Learn about architecture models, segmentation strategies, and security-first design principles.
Related Products
Other solutions you might be interested in
AdRem NetCrunch
AdRem Software
AdRem NetCrunch: agentless network and IT infrastructure monitoring. 680+ monitoring targets, automatic L2 topology maps.
baramundi Management Suite
baramundi
baramundi Management Suite: UEM platform with patch management for 200+ applications. IT automation, MDM, inventory. German manufacturer.
Citrix DaaS
Citrix
Citrix DaaS: Desktop as a Service with cloud management. HDX protocol, multi-cloud resources, auto-scaling, subscription model.
Citrix Endpoint Management
Citrix
Citrix Endpoint Management: UEM for all devices. MDM, MAM, containerization, BYOD support, Zero Touch enrollment.
Want to Reduce IT Risk and Costs?
Book a free consultation - we respond within 24h
Or download free guide:
Download NIS2 Checklist