Water intrusion and fire lead to Google Cloud outage 

Just last week Google Cloud experienced a significant outage in Paris, that spread from a single zone outage to a region-wide outage affecting over 100 cloud services. This event, which was posted to the French Network Operations Group mailing list, was the result of a cooling system water pump failure that ultimately led to a fire in the europe-west9 datacenter.  

Unfortunately, there's no estimated time for full recovery in europe-west9-a yet, and some services are still down. Google Cloud is advising its customers to failover to other regions if their data has been impacted. The event has resulted in a "multi-cluster failure that has led to an emergency shutdown of multiple zones", which explains why so many services throughout the region were affected.

The outage has understandably caused a lot of criticism about Google Cloud's disaster recovery and resilience capabilities. After all, many businesses rely heavily on cloud services for critical operations, and extended outages can cause major disruptions and financial losses. Physically separating availability zones would have prevented damage to a single zone from affecting the entire region, so it's understandable why customers are frustrated.

Google Cloud has responded by working hard to restore services and provide frequent status updates to customers. However, this incident once again highlights that the cloud is not immune to disasters, availability zones do not deliver full resilience, and an out-of-region Disaster Recovery environment is needed for business-critical databases and applications. This ensures that even when multiple zones fail, applications and databases remain available.

Stay one step ahead during a cloud outage with an out-of-region standby database 

To avoid losing revenue, customers, and damaging your good name in the market, creating a remote standby environment is mission-critical. Understanding any limitations and critical failure points is the first step. This means accepting that downtime does occur and developing a robust solution that mitigates all foreseeable risks is needed. Fortunately, software solutions like Standby MultiPlatform make it easy and efficient to create a warm standby environment that can ensure a fast recovery and near-zero data loss from any disaster. 

Out-of-region replication to a warm standby is recommended by the majority of Cloud Service Providers as the foundation of best-practice DR for business-critical systems. With StandbyMP's low bandwidth architecture, users can also easily create and maintain a warm out-of-region standby database in the cloud, benefitting from the added security, low CapEx, and ease of deployment of a cloud-based standby.