Upgrading to VCF 9.1 from vSphere + Aria + HPE Storage: The Storage-Specific Migration Guide

This is Part 13 of an architect’s read of VCF 9.1, and the fourth and final vendor-specific migration guide. Parts 10, 11, and 12 covered NetApp, Pure, and Dell. This one is for customers running HPE storage with vSphere + Aria today.

HPE’s storage portfolio under VMware has shifted significantly. The traditional 3PAR-and-Primera world has evolved into the Alletra family, with the new Alletra Storage MP B10000 as the disaggregated scale-out flagship. HPE GreenLake provides the consumption-based platform layer. Data Services Cloud Console (DSCC) is the cloud-native management plane. HPE Zerto (acquired in 2021) is the DR platform. Veeam has deep integration with HPE arrays for backup. Peer Persistence handles synchronous active-active for stretched clusters — with a key VCF constraint we’ll cover.

This piece is the architect’s walkthrough of the migration for HPE shops.

Source state: HPE storage + vSphere today

Common source-state archetypes for HPE customers:

• HPE Alletra Storage MP B10000 — newest, disaggregated, all-NVMe block scale-out, GreenLake-managed

• HPE Alletra 9000 — mission-critical block, all-NVMe, Peer Persistence-capable

• HPE Primera — predecessor to Alletra 9000, Peer Persistence-capable, still in many estates

• HPE 3PAR — legacy mission-critical platform, being phased out, end-of-life trajectory

• HPE Alletra 5000 / 6000 — mid-range, Nimble-derived architecture, cloud-managed

• HPE Nimble Storage — mid-range, predecessor to current mid-range Alletra

• HPE StoreOnce — protection storage with Catalyst integration

• HPE Zerto — continuous data protection and DR orchestration

• Veeam Backup & Replication with HPE storage integration

• HPE Data Services Cloud Console (DSCC) — cloud-native management for Alletra family

• vSphere 7.x / 8.x with vCenter, ESXi hosts, vSphere Lifecycle Manager

• Aria Operations, often with HPE InfoSight integration for storage analytics

What VCF 9.1 changes for HPE customers

Validated VCF 9.0 deployment on HPE ProLiant + Alletra MP. HPE publishes Firmware and Software Compatibility Matrices for VCF 9.0 on HPE ProLiant Gen11 and Gen12 servers using either vSAN or external Fibre Channel storage with HPE Alletra Storage MP B10000. This is the formally validated combination. For HPE customers building new VCF 9.x infrastructure, ProLiant Gen11/Gen12 + Alletra MP B10000 is the documented reference architecture.

VMFS on FC is supported as primary storage for the Management Domain. VCF 9.x removed the vSAN-only Management Domain requirement. HPE customers can deploy the Management Domain natively on an FC LUN from any HPE block array — Alletra MP B10000, Alletra 9000, Primera, or 3PAR. No vSAN required. This is consistent with the NetApp / Pure / Dell story and matters most for HPE customers who never bought into vSAN.

Peer Persistence has a vVols restriction. HPE Peer Persistence delivers transparent active-active synchronous replication between two Primera or Alletra 9000 arrays — the classic HPE stretched cluster solution. Under VCF: Peer Persistence is supported for VCF primary workload domain or supplemental storage — but only with VMFS Fibre Channel. Peer Persistence is not supported with vVols under VCF. iSCSI VMFS can be connected manually as supplemental storage. The HLD implication is direct: if you want active-active stretched clusters with HPE storage and VCF, design for VMFS-FC.

vVols deprecation impacts HPE customers. Same VCF 9.3 termination timeline applies. HPE customers using vVols — typically on Primera, Alletra 9000, or Alletra Storage MP — need a migration plan. The Peer Persistence vVols restriction (above) means many HPE customers were already on VMFS-FC for stretched clusters anyway, but standalone vVols deployments need migration to VMFS-FC, NFS, or NVMe-oF where supported.

Peer Motion enables non-disruptive data migration. HPE’s Peer Motion supports system-to-system migration without a third-party tool — for moving data from 3PAR or Alletra 9000 or Primera into Alletra Storage MP B10000. This is HPE-native, managed directly from the HPE GreenLake Cloud for Alletra MP. For customers consolidating onto Alletra MP as part of a VCF refresh, Peer Motion is the right migration tool.

DSCC is the cloud-native management plane. HPE’s Data Services Cloud Console (DSCC) provides AI-driven cloud management across the Alletra family — Block Storage, File Storage, Data Ops Manager. Cross-stack analytics give end-to-end visibility for VM environments. New AI-driven performance reporting (announced 2026) extends this further. DSCC sits alongside VCF Operations — storage-side management in DSCC, VCF-side management in VCF Operations.

Migration paths for HPE customers

Path A: Alletra MP + Convergence on FC. Customers with Alletra Storage MP B10000 (or planning to migrate to it) using VMFS on FC. vCenter and ESXi upgrade to v9, convergence brings vCenter into VCF as the Management Domain on the existing Alletra MP LUN. Datastores remain mounted. DSCC continues to provide storage-side management. The HPE-validated reference architecture for VCF 9.

Path B: Primera / Alletra 9000 + Convergence with Peer Persistence preserved. Customers with Primera or Alletra 9000, using VMFS-FC, with Peer Persistence configured for stretched cluster. Convergence path works cleanly. Peer Persistence relationships are preserved at the array layer. The VCF stretched cluster design must align with Peer Persistence’s requirements (link, latency, quorum witness).

Path C: Brownfield Import with mixed protocols. Customers with iSCSI VMFS supplemental storage, or non-vVols mixed-protocol configurations. Deploy a new VCF 9.1 instance, then import the existing HPE-backed vCenter as a workload domain. HPE’s standard supplemental iSCSI integration is preserved.

Path D: Alletra MP refresh via Peer Motion. Customers consolidating from 3PAR, Primera, or Alletra 9000 onto Alletra Storage MP B10000 as part of the VCF refresh. Use HPE Peer Motion to migrate data between storage systems non-disruptively. Then build VCF 9.1 on the consolidated Alletra MP infrastructure. Useful when storage refresh is on the same project timeline.

Path E: Side-by-side with vVols migration. Customers with significant vVols footprint. Build VCF 9.1 fresh on the chosen primary datastore type (VMFS-FC). Provision new datastores. Migrate workloads via Storage vMotion from vVols. Decommission the vVols configuration. Most HPE customers don’t have vVols-deep deployments, but where they exist, this is the cleanest path.

The Peer Persistence + Peer Motion architectural picture

HPE’s two replication/migration capabilities serve different purposes:

Peer Persistence. Synchronous active-active replication between two HPE arrays (Primera or Alletra 9000). Provides transparent failover for stretched clusters. ESXi sees a single datastore that’s actually replicated across two arrays. Requires VMFS Fibre Channel under VCF — vVols is not supported with Peer Persistence + VCF combination. For HLD: this restriction shapes the storage protocol decision when stretched cluster is a requirement.

Peer Motion. Data migration tool that moves data between HPE storage systems non-disruptively. Used for storage refresh — e.g., migrating from 3PAR or Primera into Alletra Storage MP B10000. Managed from HPE GreenLake Cloud. Not a replication tool — it’s a one-time migration mechanism that runs in the background while workloads continue to operate.

Both work cleanly with VCF 9.x as long as the protocol constraints (VMFS-FC for Peer Persistence) are honoured. For HPE customers building VCF designs, both should be inventoried and either preserved (Peer Persistence) or used as a migration tool (Peer Motion).

HPE Zerto for DR under VCF 9.1

HPE acquired Zerto in 2021. Zerto provides continuous data protection (CDP) and DR orchestration for VMware, with the architectural advantage of being storage-agnostic — it doesn’t care which array sits underneath. For VCF customers:

• Zerto Virtual Replication Appliances (VRAs) run on ESX hosts, replicating per-VM with seconds-level RPO

• Cross-site replication between VCF 9.1 instances

• Multi-target replication (e.g. production → DR site + cloud)

• Test-failover and live-failover orchestration with journal-based recovery

• Ransomware recovery via Zerto’s journal-based point-in-time recovery

• Long-term retention to object storage (including potentially VCF 9.1’s Native S3 in Part 6)

Zerto is a strong fit alongside HPE storage + VCF 9.1, particularly for customers who want consistent DR semantics across HPE arrays, vSAN, and cloud destinations. It sits alongside HPE storage-native replication (Peer Persistence, Remote Copy) and VCF’s vSAN Protection (Part 6) — different tools for different protection tiers.

Backup integration with HPE customers

HPE customers typically use one of two patterns:

Veeam Backup & Replication. Deep integration with HPE Alletra Storage MP, Alletra 9000, Primera, and 3PAR via HPE Web Services API Server. Application-aware backup, snapshot-based restores, integration with HPE StoreOnce as a backup target with Catalyst protocol. Veeam is one of the strongest backup ecosystems for HPE storage; the integration continues under VCF 9.1.

HPE Zerto. Continuous data protection alongside DR. While positioned primarily as a DR tool, Zerto’s journal-based recovery is effectively a backup mechanism with very short RPO and good ransomware recovery capability.

HPE StoreOnce. Purpose-built protection storage with Catalyst (HPE’s deduplication-aware backup transport). Continues to be the target storage for Veeam-driven backup workflows. Integrates with HPE GreenLake.

All three coexist cleanly with VCF 9.1’s native protection capabilities (vSAN Protection, ACC + IRE) — layer them by what they do best.

HLD design patterns for HPE + VCF 9.1

Pattern 1: Alletra Storage MP B10000 reference architecture. HPE ProLiant Gen11 or Gen12 servers, Alletra Storage MP B10000 for block storage, VMFS on FC for both Management Domain and workload domains. DSCC managing the storage layer. VCF Operations managing the VCF fleet. HPE-validated reference architecture for VCF 9.0/9.1.

Pattern 2: Alletra 9000 with Peer Persistence stretched. Alletra 9000 arrays at two sites with Peer Persistence between them. VMFS on FC datastores. VCF 9.1 stretched cluster design aligned with Peer Persistence requirements. Suitable for mission-critical workloads needing transparent active-active across sites.

Pattern 3: Primera + Peer Persistence (existing estate). Existing Primera deployments with Peer Persistence. Convergence path preserves the storage architecture. Eventually consider Peer Motion-based migration to Alletra Storage MP B10000 for the next generation, but the existing design works under VCF 9.1.

Pattern 4: 3PAR migration via Peer Motion. Existing 3PAR estate approaching end-of-life. Acquire Alletra Storage MP B10000. Use Peer Motion to migrate data non-disruptively from 3PAR to Alletra MP, in parallel with the VCF 9.1 migration. Eventually retire 3PAR.

Pattern 5: Mid-range Alletra 5000/6000 or Nimble. For smaller VCF deployments or branch sites, mid-range HPE storage (Alletra 5000/6000 or Nimble) with iSCSI or FC. Same VCF mechanics, smaller-scale footprint. InfoSight providing predictive analytics at the storage layer.

Common HLD/LLD anti-patterns

Designing Peer Persistence on vVols. Not supported with VCF. The combination of Peer Persistence + vVols + VCF is explicitly excluded. Stretched-cluster design must use VMFS-FC.

Trying to keep 3PAR in production indefinitely. 3PAR is on an end-of-life trajectory. While it works under VCF 9.x with VMFS-FC, it’s not the strategic direction. Plan migration to Alletra Storage MP B10000 (or Alletra 9000 for the mission-critical replacement) via Peer Motion.

Forgetting the HPE InfoSight + Aria Operations transition. HPE InfoSight provides storage-level predictive analytics. Aria Operations had an HPE management pack. With VCF Operations 9.x, validate the management pack compatibility — use Marketplace or Management Pack Builder if a fresh pack isn’t available.

Skipping the DSCC integration. DSCC is HPE’s cloud-native management for the Alletra family. Storage-side observability and lifecycle live here. Don’t try to do everything through VCF Operations — use DSCC for storage and VCF Operations for the VCF fleet.

Underestimating Peer Motion duration for large migrations. Peer Motion is non-disruptive but takes real time for large datasets. Plan the migration window appropriately and validate progress.

Treating Zerto as just backup. Zerto is continuous data protection + DR orchestration with seconds-level RPO. Use it for DR; layer Veeam + StoreOnce on top for traditional backup; use VCF’s native protection (vSAN Protection, ACC + IRE) where appropriate.

Ignoring HPE Cyber Recovery / Vault patterns. HPE has its own ransomware-resilience portfolio around Zerto + storage immutable snapshots + air-gapped archive. Map these to VCF’s Advanced Cyber Compliance (Part 1) for a complete recovery story — don’t duplicate, layer.

Skipping HPE GreenLake consumption model evaluation. For customers who want VCF without managing the underlying HPE hardware procurement and lifecycle, HPE GreenLake provides a consumption model. Worth evaluating during VCF 9.1 planning, particularly for customers who’ve been considering opex-style infrastructure.

Decision framework: HPE customer paths

• Existing 3PAR estate, end-of-life trajectory → Peer Motion to Alletra MP + VCF 9.1 Convergence on the new storage (Path D)

• Alletra 9000 or Primera with Peer Persistence VMFS-FC stretched → Convergence preserves the design (Path B)

• Alletra 9000 / Primera with Peer Persistence on vVols → Re-platform to VMFS-FC before or during VCF migration

• Greenfield VCF 9.1 deployment → ProLiant Gen11/Gen12 + Alletra Storage MP B10000 reference architecture (Path A)

• Mid-range estate (Alletra 5000/6000 or Nimble) → Convergence with VMFS-FC or iSCSI supplemental storage

• Significant Zerto investment for DR → Preserve Zerto, layer alongside HPE array-native replication

• Veeam + StoreOnce backup workflow → Preserve, validate compatibility with VCF 9.x

• HPE GreenLake consumption model interest → Evaluate GreenLake for Block Storage as the procurement path

Closing

HPE + VCF 9.1 is a clean combination once the Peer Persistence VMFS-FC constraint and the vVols migration are understood. The Alletra Storage MP B10000 + ProLiant reference architecture gives HPE customers a validated path. Peer Motion handles legacy-to-Alletra-MP storage migration without third-party tools. Zerto provides storage-agnostic DR. Veeam + StoreOnce provides backup. DSCC sits alongside VCF Operations as the storage-side management plane. The architectural decisions sit in the storage refresh strategy, the stretched cluster topology, the protocol mix, and the protection layering.

For my own design reviews with HPE customers, the questions I work through every time:

• Which HPE storage products are in scope — Alletra MP B10000, Alletra 9000, Primera, 3PAR, Nimble, mid-range Alletra?

• Is 3PAR or Primera reaching end-of-life, and is Peer Motion to Alletra MP part of the project?

• Is Peer Persistence in use — and is it on VMFS-FC (VCF-supported) or vVols (not VCF-supported)?

• vVols footprint: VCF 9.3 termination timeline considered in the migration plan?

• DSCC adoption — storage management plane configured?

• HPE Zerto adoption — DR strategy layered with array-native replication?

• Veeam + StoreOnce backup workflow — compatibility with VCF 9.x verified?

• HPE GreenLake consumption model interest — part of the procurement story?

• HPE InfoSight + VCF Operations integration — management pack compatibility?

Get those answered and the HPE + VCF 9.1 HLD writes itself.

That closes Part 13 — the HPE-specific overlay on the VCF 9.1 migration journey, and with it the four-part vendor migration mini-series (NetApp, Pure, Dell, HPE). Together with Parts 1–9, this is the 13-part read of VCF 9.1: the platform architecture, the workload class for AI, the customer migration journey, the operating model, and now the storage-vendor-specific migration recipes that get individual customers from their current estate onto VCF 9.1.

Sources

• HPE Firmware and Software Compatibility Matrix for VCF 9.0 on HPE ProLiant Gen11 with HPE Alletra Storage MP B10000

• HPE Firmware and Software Compatibility Matrix for VCF 9.0 on HPE ProLiant Gen12 with HPE Alletra Storage MP B10000

• Implementing vSphere Metro Storage Cluster (vMSC) using HPE Primera or Alletra 9000 Peer Persistence (Broadcom KB)

• HPE GreenLake Launches New Alletra MP Storage (Nexstor)

• HPE Alletra Storage MP B10000 Product Page (HPE)

• Adding HPE Alletra Storage MP B10000, 9000, Primera, 3PAR (Veeam User Guide)

• HPE Morpheus VM Essentials with Alletra MP Integration Pack

• HPE Alletra MP Storage Integration (Morpheus Docs)