Post A6usfb2sOKLwVKCF7I by tek@freeradical.zone
(DIR) More posts by tek@freeradical.zone
(DIR) Post #A6usfaelpx4RIYuzXk by admicos@mastodon.social
2021-05-04T19:20:30Z
8 likes, 9 repeats
STOP SCALING UP- SOFTWARE SHOULD NOT SUPPOSED TO SCALE- YEARS OF SCALING yet NO REAL WORLD USE FOUND for going higher than YOUR FAMILY- Wanted to go higher anyway for a laugh? We had a tool for that: It was called "EVERYONE HAVING THEIR OWN SERVERS"- "Yes please give me 5 instances of Redis. Please cluster my databases in Kubernetes Docker Container Pods" - Statements dreamed up by the utterly deranged
(DIR) Post #A6usfb2sOKLwVKCF7I by tek@freeradical.zone
2021-05-04T20:11:43Z
0 likes, 0 repeats
@admicos Also, don't underestimate how far a single PostgreSQL or SQLite instance can scale *if you're querying it efficiently*. Throwing 100 parallel servers at something doesn't do jack if your main algorithm is O(2^n).
(DIR) Post #A6usfc7WOXf9q1NSzo by admicos@mastodon.social
2021-05-04T19:21:06Z
0 likes, 0 repeats
"Hello I would like to deploy 5 instances of my frontend please"They have played us for absolute fools
(DIR) Post #A6usfddojx5gYTUlyS by admicos@mastodon.social
2021-05-04T19:22:54Z
0 likes, 0 repeats
you can tell i do not know how scaling works from the terrible examples in this jokeall i know is that if you do software with scaling up in mind it ends up terrible for my use cases and i will make jokes like this until you stop
(DIR) Post #A6usffAT42nnI1mMVM by admicos@mastodon.social
2021-05-04T19:52:43Z
0 likes, 0 repeats
Now on Gemini if you feel like sharing terrible jokes that waygemini://ebc.li/joke/stop-scaling-up.gmi
(DIR) Post #A6uvYsyHe1bzw5XXFI by smooky@pleroma.soykaf.com
2021-05-04T20:44:11.925051Z
0 likes, 0 repeats
@admicos everyoen shd have their own website and commnicate throu irc and muds
(DIR) Post #A6vl19d4Or9vMzr1mq by ssafar@fosstodon.org
2021-05-05T06:20:38Z
0 likes, 0 repeats
@tek @admicos this.There is still something I don't understand about the fact that we seem to consistently have needed expensive racks of servers for the same CRUD apps in the past 30 years, even while said servers were getting about a gazillion times faster
(DIR) Post #A6whc7SrCp26iDPMZc by tek@freeradical.zone
2021-05-05T17:17:16Z
0 likes, 0 repeats
@ssafar @admicos Well, because we need to make it highly available, of course! (Then we put it behind a load balancer, aka "single point of failure".)