Customer story · Automic and VMware Cloud Foundation

Disaster Recovery in 80 Minutes: What a Global Bank Automated

Case study recap · Broadcom

Two glossy white tiles on glass pedestals joined by a glowing arc of teal and amber light

At a Glance:

  • A global bank cut mean time to recovery by 75% with Automic Automation and VMware Cloud Foundation (VCF)
  • Core banking now fails over in 80 minutes and fails back to the primary site in 40
  • 9 out of 10 recovery tasks run with no manual intervention, and the execution logs double as audit evidence
  • Five questions to ask about your own disaster recovery runbook

Nobody schedules a disaster. Yet plenty of recovery plans read as if someone will have the whole afternoon free to work through the runbook, one step at a time.

Broadcom has published a disaster recovery automation case study about a global bank that stopped working that way. It’s a short read, and if you run Automic, it’s a good look at what Automic can do on the worst day of your year.

Recovery time

75%

Cut in mean time to recovery, per Broadcom’s case study

Failover

80 min

Core banking failover to the secondary site, with failback in 40

Hands-off

9 of 10

Recovery tasks with no manual intervention

The Problem: Manual Runbooks, Strict Regulators

The bank is a global institution processing millions of transactions a day. According to Broadcom, it faced strict regulatory mandates to recover rapidly from disasters and cyber threats with zero data loss.

Its estate is hybrid: VCF alongside Linux, Windows, and Solaris. Failovers ran on manual runbooks, and Broadcom says that stretched mean time to recovery from hours to days, with the risk of severe financial penalties and reputational damage.

What the Bank Automated

The bank made Automic Automation the control plane for failover, working alongside VCF. Three changes did the heavy lifting:

  • Policy workflows instead of static runbooks. Automic orchestrates application shutdowns, routing, and system checks as one automated sequence.
  • Readiness checks before every handoff. Automic checks system thresholds before it triggers the next dependent task, so each step waits until the systems it depends on are ready.
  • One workflow across old and new. The same automation covers the VCF private cloud and the bank’s bare-metal servers.

The Results

  • 75% faster recovery. Mean time to recovery dropped by a factor of four.
  • 80-minute failover. Core banking switches to the secondary site in 80 minutes and returns to the primary site in 40.
  • 90% automated. Nine out of ten recovery tasks need no manual intervention.
  • Audit-ready by default. Automated execution logs gave the bank immediate evidence for compliance.

“Automating our disaster recovery from end-to-end transformed a complex site failover into a reliable 40 to 80-minute procedure.”

VP of Data Center & Infrastructure Operations at the bank, as quoted in Broadcom’s case study

These are Broadcom’s published figures. The case study doesn’t name the bank, so we don’t either.

It's Not Just One Bank

Disaster recovery came up at Broadcom’s Automation Virtual Summit in September, too. Dave Kellermanns, Global Solution Architect at Broadcom, made a point worth repeating: VM-level recovery tools are excellent for bit-level copies, but they can’t restart applications or validate data, and that’s what it takes to restore a business service.

Dave shared a different story, a top-tier digital bank that encoded its entire DR runbook in Automic and took service restoration from several hours to under two minutes. In a live poll that morning, 38% of attendees picked business resilience and disaster recovery as the use case that matters most to them over the next 12 months. Our recap has the details: Intelligent Control Plane in Practice: 4 Automic Automation Use Cases.

Five Questions to Ask About Your Runbook

You don’t need a regulator breathing down your neck to get value from this. Ask your team:

  1. When did we last run a full failover test, start to finish, and how long did it take?
  2. How many steps still need a person to type a command or click a button?
  3. Does anything confirm a system is ready before the next step starts, or do we wait and hope?
  4. If an auditor asked tomorrow, could we show exactly what ran, when, and in what order?
  5. Does the runbook cover every platform, including the older servers nobody likes to talk about?

If any answer made you wince, that’s your starting point. The same idea shows up in another Automic story, Evertec’s VM rightsizing workflow: validate first, act only when it’s safe, and keep a record of every run.

Read the Case Study, Then Bring Us Your Runbook

Start with Broadcom’s case study: Global Bank Cuts Disaster RTO for Hybrid Infrastructure by 75%.

Want a second set of eyes? Bring your current runbook to a free 30-minute call with our team. We’ll look at which steps are good candidates to automate first. Just a straight conversation. Book a Free Call.

Is your runbook ready for a bad day?

RMT helps Automic teams turn manual recovery steps into tested workflows, starting with a look at the runbook you have today.