Re:Tesla engineers were on-site to evaluate the Twitter staf...
by _vertigo on Saturday October 29, 2022 @07:59PM
(#33389217)
I think you're massively underselling how complex it is to run something at scale. It's not a matter of just making the database go fast. Just using the database as an example (even though there's way more to it than just the database), you can scale a database a bit by sharding or running it as a cluster or what-have-you, but you can only shard so wide before you run into other problems which are much harder to solve. Running at a massive scale requires rearchitecting almost everything that came before.
Re:Tesla engineers were on-site to evaluate the Twitter staf...
by jvm___ on Saturday October 29, 2022 @07:55PM
(#33389189)
Twitter is just a ticket booth. Had over request, get tickets back.
Except the ticket booth is in the middle of Times Square in New York and it needs to serve millions of people per second and be instantly accessible around the globe.
The complexity is in the infrastructure, not in the complexity of what they're serving.
Re:Tesla engineers were on-site to evaluate the Twitter staf...
by x86x87 on Saturday October 29, 2022 @07:54PM
(#33389181)
Hahaha. No.
They switched to the JVM eons ago. Also, you are severely underestimating how difficult it is to build and run something like twitter at scale.
Would be really interesting to see the code base for sure and definitely exciting to modify a code base handling so much traffic… not sure I can recall a fail whale in many years so maybe they did something right or maybe fearing the whale set the large organization into stone code fear
Contrast that with Tesla and ever car is requesting data from Tesla servers and many many different complex interactions with much more complex engineering to accomplish a task and id imagine the Tesla engineering to be pretty impressive… but who knows it probably does not handle the same amount of read or write traffic…