Why a cyberattack on a peer made resilience non-negotiable for Sands

When a massive cyberattack took down the IT operations of a major peer in the hospitality and gaming industry, the lesson for Las Vegas Sands was immediate and unambiguous: in this industry, downtime means front-page news, lost revenue by the minute, and shaken customer trust.

Sands' leadership made a decision. Resilience wouldn't be a document sitting in a drawer waiting for a disaster. It would be baked into the cloud itself.

How manual runbooks and partial IaC coverage left Sands exposed

Like most large enterprises, Sands' recovery posture had grown organically, and it showed:

  • Manual runbooks carried the weight of the recovery strategy. Executing them meant days of coordinated engineering effort, with plenty of room for human error under pressure.
  • IaC coverage was partial. Significant parts of the environment existed only in the cloud, with no code to rebuild them from.
  • Inventory was an in-house project: expensive to maintain, hard to keep current, and never quite complete. (You can't recover what you can't see.)

A real DR event would have meant reconstructing infrastructure from memory, tickets, and tribal knowledge: exactly the scenario that crippled their peers.

How Firefly turned resilience into a property of the infrastructure itself

Sands chose Firefly to turn resilience from a manual process into a property of the infrastructure itself:

That included complete, real-time inventory across the entire cloud footprint: every resource, managed or not, visible in one place. It also included automated codification of unmanaged resources, converting the live environment into versioned Infrastructure-as-Code. Plus, automated DR drills and failover, replacing runbooks with repeatable, testable recovery workflows.

The results: 96.4% codification, DR failover under 20 minutes

Within the engagement, Sands reached 96.4% codification coverage of their cloud infrastructure: meaning nearly the entire environment can be rebuilt from code, on demand.

Just as importantly, DR testing went from a dreaded, days-long exercise to a routine one: Sands can now automate a full DR drill and fail over to a different region in under 20 minutes.

Resilience is no longer Sands' contingency plan. It's how their cloud is built.