Apollo’s hosted services such as Uplink and cloud routing are designed and maintained by the best in the business using the best infrastructure possible. Our primary cloud provider is Google Cloud with failover to Amazon Web Services for critical services. For our hosted Serverless offering, we are partnering with Fly.io to ensure not only high availability for routing traffic, but the ability to be as close to the edges of our clients as possible.

Apollo services are available a minimum of 99.9% of the time, measured on a monthly basis, excluding scheduled downtime. You can view all of these at any time by visiting Apollo’s Status Page. If an incident is declared, updates are given every 15 to 30 minutes on average.

The concept of Fail Open is a core principle at Apollo: A system that fails is designed not to shut down when failure conditions occurs. Instead, the system remains open and operations continue as if the system were not even in place. Our Blue/Green deployments work the same way. All infrastructure is written as code to provide insight, baselines, auditability, and review. View our live Status Page here.

Emergency repairs are part of the business of supplying software as a service. We do annual disaster recovery drills as part of our SOC 2 audit. Scheduled downtime is kept to a minimum, communicated well in advance with customers, and rehearsed down to the level of what topping goes on the pizza everyone will share during the event. We don’t want our service down any more than our customers.
Please contact your Apollo Customer Success and Solution Architect teams to discuss resilience scenarios and mitigation strategies for your organization.