November 6, 2025 Implementing Unified SQL Server HA/DR in Manufacturing

Keep your facility running with unified HA/DR for SQL Server – boost uptime, cut costs, and protect production data.

By Don Boxley, CEO and Co-Founder, DH2i

I am likely not shocking you with the news that in the manufacturing industry, even the shortest of system outages can throw everything off – from production halts, to orders getting backed up (or lost), to very unhappy customers (and thereby your organization’s management). Like virtually every industry today, manufacturers depend on data. Automated machinery and robotics to IoT sensors and ERP systems – everything is generating and depending upon data. Which, of course, makes it absolutely critical to keep SQL Server databases running smoothly and recoverable at all times, as SQL is the foundation on which most of these apps sit.

Putting together a strong high availability (HA) and disaster recovery (DR) strategy isn’t just some checklist exercise. For IT and data center pros, it’s about building something that actually holds up in the real world. Done right, it keeps things running, streamlines your stack, and protects your most critical systems from surprise outages. And let’s be honest…you already know what happens if it’s not done right…

Key Challenges in Manufacturing HA/DR

Minimizing Downtime on the Factory Floor
A global electronics manufacturer learned the hard way just how costly downtime can be. A power surge at one of their plants took the MES database offline for almost an hour. Without HA or DR in place, order fulfillment came to a standstill and the delays quickly spread to three other production sites.

Controlling Cost While Expanding Reliability
A precision tool maker with five plants spread across the United States needed to boost uptime but couldn’t justify the cost of high-priced third-party replication software. Instead of overhauling everything, their team leaned into what they already had i.e., SQL Server’s native features, and made the switch to Linux. The result was impressive: they hit 99.99 percent uptime and brought operating costs down by nearly a third.

Adapting to Hybrid and Multi-Site Environments
A consumer goods manufacturer was struggling to get DR right between its Midwest and Southeast locations. One site ran on Windows, the other on Linux, and trying to get the two to play nicely was a headache. Instead of ripping everything out and starting from scratch, they needed a flexible solution that could work with what they already had — something that could bridge the gap and pull it all together without a full-blown overhaul.

Step 1: Plan Your HA/DR Architecture Based on Operational Priorities

Start by categorizing applications and workloads based on their criticality to production and delivery timelines.

Example Tiers (with scenarios):

• Tier 1: Multi-Site HA + DR
  • Use case: A global automotive parts supplier maintains an MES and SCADA system that supports 24/7 manufacturing across two continents
    They implemented HA clusters within each plant and AG-based DR across regions to avoid downtime from either local server failure or regional disasters.

• Tier 2: Multi-Site AG Without HA
  • Use case: Packaging manufacturer runs its product analytics and QA systems at two plants
    They set up Availability Groups (AGs) across both locations, allowing them to maintain continuous insight into production quality without investing in full HA setups.

• Tier 3: Single-Site HA Only
  • Use case: Local furniture maker uses a lightweight ERP system and inventory control at a single site
    Windows Server Failover Cluster (WSFC) ensures these systems stay online during hardware maintenance or server issues.

Step 2: Choose Platform-Agnostic, Scalable Technologies

While preparing for future transitions to cloud or Linux-based infrastructure – use tools and architectures that support your current mix of systems.

• Support for Windows and Linux
  • Case example: Metal parts supplier gradually transitioned from Windows-based SQL Server to Linux-based clusters after realizing the licensing savings
    Their IT team reported 25% lower TCO over 18 months while maintaining full compatibility and performance.

• Extended Virtual Hosts (Vhosts)
  • Case example: Food manufacturing company deployed extended Vhosts across two cold-storage facilities
    This allowed them to run HA for local operations (e.g., temperature monitoring systems) while replicating critical data for DR between sites, without third-party tools.

Sample Deployments:

1. Two geographically separate facilities
   Plant A and Plant B (e.g., Georgia and Texas) each run two SQL Server nodes to ensure local HA.
2. At each location, two SQL Server instances with shared storage
   HA is maintained using WSFC or Pacemaker (Linux) clusters.
3. AG replication between sites
   Ensures cross-site DR for systems like centralized inventory and shipping databases.

Step 3: Implement Based on Tier

Follow through on the architecture based on what each facility or application requires.

Tier 1: Multi-Site HA + DR

• Deploy HA-capable SQL Server instances at each facility
  Use clustering or failover mechanisms to ensure critical apps like MES and IoT gateways never go down locally.

• Enable AG replication for cross-site data sync
  Maintain real-time backups and synchronization to a remote site in case of physical damage or ransomware.

• Configure automatic failover within each site and manual failover across sites
  A beverage manufacturer practiced monthly DR failovers between its Midwest and Southwest plants to ensure operational continuity.

Tier 2: Multi-Site AG Without HA

• Use standalone SQL Server instances at each site
  For manufacturing analytics and performance dashboards not tied directly to shop-floor control.

• Set up AG replication between the sites
  Ensures continuity of reporting data even if one site experiences a network or server failure.

Tier 3: Single-Site HA

• Deploy local SQL Server cluster within a plant or office
  Used by a toy manufacturer to protect a small but critical design database shared across departments.

Step 4: Reduce Infrastructure and Licensing Costs

Lean IT budgets in manufacturing make cost efficiency a critical component of any HA/DR strategy.

• Migrate SQL Server workloads to Linux
  A medical device manufacturer achieved a 40% reduction in licensing costs by migrating from Windows Server to Linux-based SQL deployments.

• Use containers to consolidate workloads
  A semiconductor plant runs containerized SQL Server instances for shift scheduling, lab data, and order processing, cutting their SQL Server license usage in half.

• Replace third-party replication tools with SQL Server AG
  A pet food manufacturer eliminated $60,000/year in replication software fees by switching to AGs across its two main sites.

Step 5: Test and Monitor

Don’t wait for a real failure to find out if your DR plan works.

• Test both intra-site and cross-site failovers regularly
  A plastics manufacturer conducts quarterly failover drills between its plants and HQ IT center.

• Set up automated alerts and monitoring
  Ensure anomalies in performance or failover readiness are detected early—especially during production runs.

• Integrate HA/DR health into factory dashboards
  One automotive electronics company displays DR system status alongside machine KPIs so plant managers can see everything in one place.

Finished Goods

Manufacturers can keep everything humming — on the factory floor, across the supply chain, and in IT — by combining high availability and disaster recovery into one streamlined, scalable setup. Even if budgets are tight, solutions like extended virtual hosts and SQL Server AGs make it totally doable to build a rock-solid infrastructure without overspending. Whether you’re running a single plant or managing operations around the world, this kind of approach keeps you online, helps you recover quickly when things go sideways, and sets you up for whatever’s coming next — without letting costs spiral out of control.

About the Author:
Don Boxley Jr is a DH2i Co-founder and CEO. He has more than 20 years in management positions for leading technology companies. Boxley earned his MBA from the Johnson School of Management, Cornell University.

Subscribe to Industry Today