3 ms·
Seriously. It's impossible to manage systems (ops) and develop at the same time if you have any sort of complex system. With AWS and such, you need permanent op
by joesmo 11y ago
Seriously. It's impossible to manage systems (ops) and develop at the same time if you have any sort of complex system. With AWS and such, you need permanent ops to deal with all the bugs, issues, and slow support. With a solution like Docker you need someone to set that up and manage the cluster. Either way, you need ops unless your infrastructure is very basic.
- hobs 11y agoI think the reality is that it is somewhere in-between. Most companies dont want to spend on ops anyway, so they provision the least number of devops people and then complain that their deployments are slow or whatever results they are attempting are produced slowly (mostly due to emergent work due to ops struggles.) I really like the goog model of your SREs spending 50/50 dev time and ops time, because it forces a shift in how you are deploying things and the ops people have a big say in whatever the devs are doing (not the case in many companies.) It is not impossible to do any of these things, but you cant get around the basic complexity problem by saying "oh this is dead now, just pay someone else to do it." because the fact is you are giving over your keys to the kingdom. For instance, People have been saying for years that "The DBA is dead." but I work with many companies that building new database driven products, thinking they can handle concurrency and growth, and then when the real complexity starts occurring because their abstraction is a bit leaky shit hits the fan. An argument against specialization only works when you have completely replaced that industry with something entirely better (we dont have a lot of blacksmiths anymore), and that has definitely not happened for devops.