3 ms·
Just wanted to confirm something. You say you have no single point of failure but you are running everything in one data center ?
by taligent 14y ago
Just wanted to confirm something. You say you have no single point of failure but you are running everything in one data center ?
- simonw 14y agoThat was slightly loose wording in the article: yes, the single data center is currently a single point of failure (though we can spin up a replacement stack elsewhere very quickly). Andrew was referring to the way we had rearranged our software stack to reduce the SPOFs (our previous architecture had a single load balancer for example).
- twerquie 14y agoThere is always a single point of failure if you zoom out far enough.