6 ms·
I think the "single dependency" was moreso about the technology stack, not the number of instances of whatever the technology is.
by VWWHFSfQ 5y ago
I think the "single dependency" was moreso about the technology stack, not the number of instances of whatever the technology is.
- cortesoft 5y agoThey mention that one of the reasons is having fewer things that can fail and bring down your system... two separate PG systems doubles the failure chance
- voakbasda 5y agoIf you don't maintain redundancy (i.e. more than one instance), you will be 100% certain to have a service outage at some point in your future.
- cortesoft 5y agoSure, but you still have double the chance of failure.. Two distinct PG clusters you depend on are twice as likely to have an outage as one.