3 ms·
AWS had its own share of massive outage. I had a look at AWS and GCE incidents history and it looks the frequency and magnitude of incidents is quite similar. D
by ngrilly 12y ago
AWS had its own share of massive outage. I had a look at AWS and GCE incidents history and it looks the frequency and magnitude of incidents is quite similar. Do you have any factual information that contradicts this?
- plnk22 12y agoPerhaps the fact that the GCE outage affected multiple regions around the world. EC2 outages were all localized to a single region (and in most cases limited to a single AZ within that region).
- orenbarzilai 12y agoAWS asks you to design your architecture for failure and work across multiple availability zones. Its complicated and expensive but you can avoid that. With google from the other end you can't. and that's the price you pay when you choose PaaS over IaaS
- tapirl 12y agoGoogle Compute Engine is also an IaaS. You can also design your architecture for failure and work across multiple availability zones by using GCE. GCE is not GAE.
- plnk22 12y agoGCE had a multi region outage so multi zone within a region wouldn't have helped in this case.
- ngrilly 12y agoI agree. This is the reason why this outage is more serious than others.
- orenbarzilai 12y agoMy mistake, mixed GCE with GAE