Data Center Consolidation: Hub-Spoke Migration
A Philippine bank had 8 data centers across the archipelago: Manila (2), Cebu (2), Davao (1), Clark (1), Iloilo (1), and Zamboanga (1). Each was independently managed, separately licensed, and differently configured. Total annual cost: $4.2 million.
After consolidation to 3 data centers (Manila hub, Cebu hub, Clark backup), annual cost dropped to $2.1 million — a 50% reduction. But the real win wasn't cost. It was consistency: one configuration, one team, one set of processes across all sites.
Data center consolidation is painful but transformative. Here's how to do it without breaking your business.
What is Data Center Consolidation?
Data center consolidation is the process of merging multiple data centers into fewer, larger facilities. The goal: reduce costs, improve efficiency, and simplify operations by eliminating redundant infrastructure.
The hub-spoke model is the most common consolidation architecture:
Hub (1-2 sites): Large, centrally-located data centers that host most workloads. Full redundancy, enterprise-grade infrastructure.
Spoke (2-4 sites): Smaller facilities that handle local workloads (branch office applications, local data processing) and serve as DR for the hub.
Think of it like a airline hub: most flights connect through the hub, but smaller airports (spokes) serve local traffic and provide backup routing.
Why Consolidate?
The business case for consolidation is compelling:
Cost reduction: Fewer facilities = less power, cooling, rent, and staff. Typical savings: 30-50%.
Operational efficiency: One team manages everything. One set of processes. One monitoring system.
Better security: Fewer physical locations to secure. Concentrated security investment.
Improved DR: Consolidated infrastructure is easier to replicate and protect.
Scalability: Larger facilities can scale more efficiently than many small ones.
The downside: migration risk. Moving workloads between data centers is complex, time-consuming, and can disrupt operations if not planned carefully.
Hub-Spoke Architecture Design
Here's the architecture we designed for the bank:
Hub Site (Manila):
Primary data center: 200 racks, 500+ VMs, full redundancy (2N power, redundant cooling).
Hosts: Core banking, EHR, ERP, and other mission-critical workloads.
Network: 10Gbps backbone, redundant ISP connections, Direct Connect to AWS/Azure.
Hub Site (Cebu):
Regional hub: 80 racks, 200+ VMs, N+1 redundancy.
Hosts: Regional workloads, Visayas-Mindanao operations, DR for Manila hub.
Network: 10Gbps backbone, dedicated WAN link to Manila (latency <10ms).
Spoke Sites (Clark, Iloilo):
Local processing: 10-20 racks each, 20-50 VMs.
Hosts: Branch office applications, local data processing, edge workloads.
Network: 1Gbps WAN to nearest hub, local internet for cloud access.
Key design principle: spokes should be self-sufficient for local operations. If the WAN link to the hub goes down, spoke sites continue running local applications.
Migration Planning
Migration is the hardest part. Here's our 5-phase approach:
Discovery (4-6 weeks): Inventory all workloads across all data centers. Document dependencies, performance requirements, and compliance constraints. We use automated discovery tools (RVTools, Nutanix Xi, or Azure Migrate).
Design (4-6 weeks): Map workloads to the new architecture. Which workloads go to which hub? Which stay at spoke sites? What network changes are needed?
Pilot migration (4-8 weeks): Move 10-20 non-critical workloads first. Test performance, validate networking, verify backup/DR. Fix issues before the big migration.
Production migration (12-24 weeks): Move workloads in waves. Each wave = 10-20 VMs. Migrate during maintenance windows. Have rollback plans for every wave.
Total timeline: 6-12 months for a multi-site consolidation.
Migration Techniques
For VM-based workloads, we use these migration techniques:
Live migration (minimal downtime):
Nutanix: Xi Leap for cross-site migration.
VMware: vSphere vMotion for same-site, SRM for cross-site.
General: Veeam or Zerto for replication-based migration.
Cold migration (planned downtime):
Export VM, transfer to new site, import. Simple but requires downtime window.
Application-level migration:
Database replication (SQL Server Always On, MySQL replication).
File sync (rsync, DFS Replication).
DNS cut-over (update DNS to point to new location).
The key: always migrate during maintenance windows. Always have a rollback plan. Always test before cutting over.
Network Design for Consolidation
Network connectivity between hubs and spokes is critical:
Hub-to-hub: 10Gbps dedicated WAN (MPLS or leased line). Latency <10ms.
Hub-to-spoke: 1Gbps WAN. Latency <20ms.
For the bank, we used PLDT Enterprise MPLS for hub-to-hub and Globe Business for hub-to-spoke. Redundant providers for each link.
DNS planning is often overlooked. During migration, DNS TTL (time-to-live) must be set low (300 seconds) before cutover. This ensures quick failover if something goes wrong.
Cost Analysis
Here's a real cost comparison for the bank consolidation:
Before (8 data centers):
Facilities: $1.8M/year (rent, power, cooling).
Hardware: $1.2M/year (leases, refresh cycles).
Staff: $800K/year (8 teams, 24 staff).
Software: $400K/year (per-site licensing).
Total: $4.2M/year.
After (3 data centers):
Facilities: $900K/year (3 sites, optimized power).
Hardware: $600K/year (consolidated, newer hardware).
Staff: $400K/year (1 team, 12 staff).
Software: $200K/year (consolidated licensing).
Total: $2.1M/year.
Savings: $2.1M/year (50%). Migration cost: $800K (one-time). Payback period: 5.7 months.
Risk Management
Consolidation carries real risks. Here's how we mitigate them:
Data loss: Use verified backup before every migration. Test restore at the destination before cutting over.
Downtime: Migrate during maintenance windows. Have rollback plans. Accept that some migration will extend into the window.
Performance degradation: Benchmark workloads before and after migration. If performance drops, investigate immediately.
Compliance gaps: Map regulatory requirements to the new architecture. Ensure data sovereignty and retention requirements are met.
Staff resistance: People fear change. Communicate early. Explain the benefits. Involve the team in planning.
Best Practices
Start with an assessment. You can't consolidate what you don't understand. Document every workload, every dependency, every compliance requirement.
Migrate in waves. Don't try to move everything at once. Start with non-critical workloads. Build confidence. Then tackle mission-critical systems.
Invest in network. The WAN between hubs and spokes is the backbone of your consolidated architecture. Don't skimp.
Plan for the long term. Design for 5 years of growth. It's cheaper to over-build during consolidation than to retrofit later.
Communicate constantly. Migration affects everyone. Keep stakeholders informed. Set expectations for downtime. Provide status updates.
Conclusion
Data center consolidation delivers massive cost savings and operational improvements, but it requires careful planning and execution. The hub-spoke model provides the right balance of centralization and local resilience.
Start with a comprehensive assessment. Design the target architecture. Migrate in waves. Test every step. The pain is temporary — the benefits compound for years.
Want to go deeper? Explore [Run infrastructure services](/en/products/run), [industry solutions](/en/solutions), or [contact our team](/en/contact).
FAQ
Q: How long does a multi-site consolidation take?
A: For 3-5 sites: 6-9 months. For 5-10 sites: 9-15 months. The timeline depends on workload complexity, compliance requirements, and available migration windows.
Q: Can I consolidate without downtime?
A: Most workloads can be migrated with minimal downtime (under 30 minutes) using live migration. Some workloads (legacy applications, databases) may require longer maintenance windows.
Q: What if consolidation doesn't meet cost targets?
A: Revisit the design. Common cost overruns: underestimating network costs, over-provisioning the hub site, or not retiring spoke sites completely. Adjust the architecture based on actual costs.
Q: Should I use cloud instead of consolidating?
A: Cloud is a form of consolidation — you're consolidating to a cloud provider's data center. Evaluate both options: on-premises consolidation vs cloud migration. For regulated industries with data sovereignty requirements, on-premises consolidation is often the better choice.
Industry Trends and Market Analysis
The market for this technology is growing at 15-25% annually, driven by digital transformation initiatives, remote work requirements, and increasing security concerns. According to Gartner, 75% of enterprises will have deployed this type of solution by 2026, up from 35% in 2023.
Key trends to watch: cloud-native architectures are becoming the default, AI/ML integration is moving from nice-to-have to essential, and zero-trust security models are replacing perimeter-based approaches. Organizations that delay adoption risk falling behind competitors who leverage these technologies.
Vendor Selection Criteria
When evaluating vendors, focus on five key criteria: technical capability (does it meet your functional requirements?), scalability (can it grow with your organization?), support quality (what is the SLA and response time?), total cost of ownership (not just purchase price), and ecosystem (partners, integrations, community).
We recommend creating a weighted scoring matrix with these criteria. Assign weights based on your priorities (e.g., if support is critical, give it 30% weight). Score each vendor on a 1-5 scale for each criterion. The vendor with the highest weighted score is usually the best fit.
Change Management and Adoption
Technology implementation is only 50% of the project. The other 50% is change management. People resist change, especially when it affects their daily workflows. Invest in communication, training, and support to ensure adoption.
Key change management steps: identify champions (early adopters who can advocate for the new solution), provide hands-on training (not just documentation), create feedback loops (regular check-ins with users), and celebrate wins (share success stories to build momentum).