A greater need to simulate and rollback changes will resolve another Google outage.

 

Guy Bunker, chief scientist at Symantec, claimed that as the Google outage last Thursday afternoon was caused by administrator error there should be plans made overall to simulate and rollback changes.

 

Bunker claimed that as networks are becoming more complex, there is a greater opportunity for error.

 

“How many times has one of your administrators made a mistake? Probably more frequently than you would imagine - but because it doesn't effect you in the same way, you don't notice. The need to effectively simulate changes is going to become essential in cloud service providers in order to prevent such problems. The ability to quickly rollback changes is also going to be needed so that problems like this can be quickly resolved,” said Bunker.

 

Although he did claim that a ‘degradation of service is better than no service at all', the fact that human error was the cause showed that a large network can be resilient to disaster.

 

Bunker said: “If you are looking to put your data into the cloud, or use applications in the cloud, then look at the details the service providers give on disaster recovery and business continuity as well as how they plan and implement changes to the infrastructure. If you can have visibility into these things then you can plan around them.”