Top US Municipal Government
Migrates Close to 200 Legacy
Vblock Workloads to a Modern
Hybrid VMware Cloud Foundation
Platform
Key highlights
~200 workloads migrated
Moved mixed production and non-production workloads off end-of-life Vblock infrastructure.
~50 workloads kept their IPs
Retained the same IP addresses by recreating subnets as NSX-T overlay segments.
Multiple racks to one
Replaced the legacy footprint with a 9-blade VCF cluster and two top-of-rack switches.
Summary
The client needed to retire end-of-life Vblock infrastructure running unsupported vSphere 5.x and hosting close to 200 servers. The team designed a hybrid landing zone on VMware Cloud Foundation and VMware Cloud on AWS, migrated mixed workloads using Rubrik and VMware HCX, preserved IP addresses for around 50 workloads, and decommissioned the legacy gear.
Client overview
Supporting critical city operations on an aging, end-of-life IT infrastructure
The client is the administrative body of a large US city, delivering public safety, utilities, libraries, parks and recreation, and general administration services through a central office and many satellite locations, supported by a workforce numbering in the thousands.
Its legacy Vblock platform ran on end-of-life hardware and unsupported VMware vSphere 5.x, hosting nearly 200 servers that needed to move to a more reliable, compliant, and efficient infrastructure.
Challenges
End-of-support infrastructure with no vendor safety net
No vendor support: The Vblock was end-of-life and end-of-support, so hardware or software failures had no vendor backing.
Incompatible migration tools: No standard migration tool supported the version, which had been end-of-life since 2018, making a low-downtime approach hard to define.
Storage pressure: Legacy VMFS datastores did not auto-reclaim space, repeatedly filling storage pools, causing high-severity incidents, and leaving no room to add disks or restore quickly after failures.
IP address dependencies: Around 50 workloads could not change IP addresses, so their addressing had to be retained after the move.
Unknown servers: Unidentified 'mystery' servers and unclear application ownership complicated migration planning.
Solution
Hybrid cloud modernization, phased workload migration, and seamless IP preservation
Workloads were moved to a hybrid platform: VMware Cloud Foundation on-premises, built on HPE hardware under a GreenLake subscription model, and VMware Cloud on AWS. Rubrik moved workloads to on-premises VCF; where restores failed, VMware HCX moved them to the cloud in a two-step migration.
To keep IP addresses, the team moved entire subnets, removed the gateway from the old network, and created an overlay segment with the same subnet in NSX-T. Aria Network Insight traffic analysis exposed server dependencies and owners. Work followed phases of design acceptance, hardware ordering, deployment, migration, and closure, with application testing after each move group.
Hybrid landing zone
On-premises VMware Cloud Foundation on HPE hardware plus VMware Cloud on AWS, consumed through a subscription model.
Workload migration
Rubrik restores for on-premises moves and VMware HCX for cloud moves, organized into move groups with application testing.
IP retention
Subnets recreated as NSX-T overlay segments so around 50 workloads kept their existing IP addresses.
Dependency discovery
Aria Network Insight flow analysis identified application dependencies and server owners.
Business outcomes
A stable, supportable, and significantly more compact platform
All workloads moved to a more reliable, compliant infrastructure, and the legacy Vblock, which occupied multiple racks, was decommissioned and replaced with a compact VCF cluster of 9 blades and two top-of-rack switches in a single rack. High-severity incidents caused by failing disks and storage-pool reclaim issues stopped, keeping SLAs green, and virtual machines were upgraded from old VMware Tools and hardware versions. The new platform was provisioned to serve the client for at least five years before major work is needed, with the subscription model giving flexible capacity and cost control. Lessons learned included allowing more time to identify server owners and better managing downtime scheduling and dependent projects.
