0
Google Compute Engine (GCP) servers are redundant by themselves from creation?
I made my server in the Iowa area but wanted to know if something happens there, the server would continue running in another Google zone without losing the data.
Is it written somewhere that this is guaranteed by Google?
0
It has redundancy, but not by default. See some techniques:
With the GCP Load Balancer you can have its application in multiple regions.
Deploy instances Across Multiple Regions using global load Balancing. HTTP(S) Load Balancing Enables your Traffic to enter the Google Cloud Platform system at the Location Nearest the client. Cross-regional load Balancing provides Redundancy so that if a Region is unreachable, Traffic will Automatically be diverted to Another Region so that your service remains Reachable using the same External IP address.
See a full example of how to make a load between regions.
The Compute Engine has what they call Live Migration. It ensures that if something happens to the machine used, it automatically migrates to another machine in the same region.
Remembering that this happens whenever possible, because if there is a problem that prevents the live Migration, the virtual machine restarts itself and writes a category log hostError.
Also worth reading the Google Compute Engine SLA, which is 99.99% (2019).
0
To the extent that I know, if you want to ensure this type of availability, you must have at least another machine registered in another zone/region, using a load and template instance groups for control.
Suddenly you can justify to your customer the issue of SLA provided by Google in this aspect
Browser other questions tagged server
You are not signed in. Login or sign up in order to post.
What kind of data are we talking about? Database, images, documents? Maybe there are some other options within the GCP’s own backend.
– vinibrsl