Google shuts down key data center due to London’s heatwave. The tech giants have made a tactical decision in shutting down one of its data centers on the hottest day ever.
Google Shuts down Key Data Center Due To London’s Heatwave
Tech company Google has now revealed more information in regards to what happened when it was forced to shut down one of its London data centers on the UK’s hottest day of the year so far.
According to Google, the failure of zone “Europe-west2-a” last month was down to not maintaining a safe operating temperature all due to a simultaneous failure of multiple, redundant cooling systems that is combined with the “extraordinarily high” outside temperatures.
The failure impacted several Google services, and this is including Google Compute Engine, Persistent Disk (PD), and Google Cloud Storage, thus causing instance terminations, service degradation, and networking issues.
What Really Happened
Engineers at Google shut down the data center hosting most of the impacted some Europe-West2-a all the while the cooling system was repaired. The general and total impact on cloud services is estimated at 18 hours and 23 minutes.
This is somewhat disturbing news given the fact on how Google claims that these regional services are “designed to survive the failure of a single zone.”
Google however attributed the mistake to inadvertently modifying the routing of traffic for its internal services in order to avoid all three zones in the ‘Europe-west-‘region rather than the sole impacted ‘Europe-west2-a’ zone.
The resulting routing incident stopped its customers from being able to get access to data from regional storage services and this is inclusive of GCS and BigQuery across several zones.
Will This Incident Happen Again
News such as this is understandably pretty scary, that is, if you are concerned about global warming, as the UK might well be seeing warmer days in the future.
Luckily, Google has however made some commitments in its bid to stop these types of failures from impacting its cloud hosting ever again.
Some of the commitments made by the company included repairing and re-testing its failover automation in a bid to ensure stronger resilience in its failover protocols during large-scale events just like this one.
The cloud giant also is committed to investigating and developing “more advanced methods” in order to progressively decrease the thermal load that is within a single data center space, thus reducing the probability that a full shutdown is required.
Google in addition is reportedly set to examine its procedures, tooling, and automated recovery systems for various gaps and will also be conducting an audit of its cooling systems equipment and standards across its various data centers that house Google cloud all over the globe.
