Google Cloud is having a significant outage in Europe, and the damage seems to reach well past Google's own products.
The company's status page said "Multiple services are experiencing issues in the europe-west2 region" and logged the problems at 10:24am BST on October 20. A region is a cluster of data centers, so one failing means every service hosted there can stumble together. The status page showed orange alerts across many systems, including BigQuery (Google's data analysis tool), App Engine, Looker and its SQL database services. Reports describe many European regions as affected, which is broader than the single region Google named in its statement. Google has not said what caused the fault or when it expects a fix.
Outage tracker DownDetector, which counts user complaints, showed a spike for Google Cloud that dipped slightly as the morning went on. It also showed spikes for other companies that followed the same pattern, among them the banking apps Revolut and Lloyds Bank in the UK. That is circumstantial evidence, not confirmation. Neither bank had said Google was the cause, and user-report spikes can have other explanations.
Still, the pattern is the point. Most customers have no idea which cloud their bank runs on, and a provider's status page is not something they check. When a shared back end falters, the failure shows up as a broken banking app, with no hint of where the fault sits.
The useful question comes later, once Google publishes its account. If companies were hit because they kept everything in one region, that is a design choice each of them can fix. If a fault spread across several regions at once, that is a harder problem, and it belongs to Google.