[[Five 9s by 2029]] Even if we have to rewrite the whole stack.
Show this thread
-
5 9s would be cool, but few systems can justify that level of ongoing investment. For systems that don’t need it, how does one prevent the extra cost from interfering with the primary business needs?
Why does 5 9s require some kind of ongoing investment? Why don't computers just work that well by now, as a matter of basic principle? We have been working on them for 8 decades, surely they have been getting better the whole time?
-
-
The craziest thing about moving to Clojure from Ruby or JavaScript — when you find a library that hasn’t been updated in years, often it’s just because it’s DONE As long as it isn’t a wrapper of something JS, very often no need to keep updating things that just did the right.
-
Totally agree, Language, ecosystem, and dependency risk awareness are often overlooked as important differentiators. (Look at the erlang ecosystem, with their showcase of 9 9s applications built on them)
- Show replies
New conversation -
-
-
The tolerance for mistakes and scrutiny is higher to mitigate tiny outages. As a single tiny botched deploy may screw your whole years SLA. Very few “at scale” Systems with 5 or more 9s, that also are products and not simple services, also innovative and evolve rapidly.
-
There is a natural tension between “change” and “high 9s” and “affordability” Remember also, many systems are only accidentally achieving such an SLA…
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.