#fault_tolerance — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #fault_tolerance, aggregated by home.social.
-
@adastra1930 good advice! #hybrid setups provide #fault_tolerance too. If the cloud provider goes down, or trunk connections are cut, or there's a massive public #DNS outage, or even a large-scale power outage - you can still have #authentication services, and some #connectivity with internally hosted sites and services. Plus, managing an on-prem Domain Controller gives we #sysadmins the control on routes and content delivery for our users and devices, and we can pivot quickly to recover services should the Internets take a massive shit. Of course, there are limits to what can be done today without some #cloud services in place, or even #electricity, but that's not the point here.
There are fewer and fewer of us anymore that actually know how to spec and build out a #physical_server for a target environment, let alone how to make it play well with clouds. And don't get me started on #n00b tech people's lack of understanding on how the #Internet actually functions. It's a lost art form anymore lol.
I still believe A+, and Networking Fundamentals (Level 1 & 2) courses should be required learning for anyone in computer science. Part of passing those courses should be building a PC, building a Server, and building a Network, from scratch. Then break/fix/break/fix until you "get it." We also need more #support for #n00bs landing #industry jobs with quality #leaders who actually #mentor them, making them better than they could be. Teaching these now older concepts of "on-prem" setups and the like.
Now, if I could just get my CEO to release the purse strings a bit, I could continue doing my part. But that's a whole other discussion lol.
-
Проектируем микросервисы с Reactive Manifesto
Работаете с распределенными системами или только пытаетесь к ним подобраться? Проектировать такие системы с нуля бывает сложно и страшно: чтобы учесть все нюансы, нужен определенный багаж знаний. На помощь приходит Reactive Manifesto — документ, который обобщил опыт целого ряда компаний по созданию распределенных систем. Манифест формулирует главные принципы, на которые стоит опираться на этапе проектирования и эксплуатации таких систем. Меня зовут Андрей Василевский, я системный архитектор в Lamoda Tech. В этой статье я на примерах из своей работы покажу, как применять Reactive Manifesto на практике. Статья будет полезна тем, кто только начал изучать распределенные системы, хочет закрепить теорию или тем, кто хочет структурировать проектирование микросервисов в своей компании.
https://habr.com/ru/companies/lamoda/articles/803793/
#микросервисы #domain_driven_design #event_driven_architecture #system_design #architecture #saga #распределенные_системы #cap_theorem #fault_tolerance #guideline
