Ancestors

Toot

Written by froztbyte@awful.systems on 2024-09-09 at 12:38

ibm mainframes: now featuring AI

https://awful.systems/post/2338432

=> More informations about this toot | More toots from froztbyte@awful.systems

Descendants

Written by froztbyte@awful.systems on 2024-09-09 at 12:45

the press release (archive) says:

featuring its first advanced on-processor chip AI accelerator for inferencing

For instance, our AI-driven fraud detection solutions are designed to save clients millions of dollars annually. With the introduction of the AI accelerator on the Telum processor, we’ve seen active adoption across our client base. Building on this success, we’ve significantly enhanced the AI accelerator on the Telum II processor

if I’m reading this correctly, it’s on-die in the telum ii, but was a separate thing (like co-processor or architecture add-in card or something) previously?

the usecase sooooort of makes sense but I’m still skeptical about part of it because this seems awfully like it’d be potentially limited by changes over time in how one might do such tasks (e.g. if a new preferred inferencing method comes out that doesn’t quite fit the chip pattern). but also “our AI-driven fraud detection solutions” - ah.

guess it’ll be interesting to see how this shit sits in 10y or something.

=> More informations about this toot | More toots from froztbyte@awful.systems

Written by skillissuer@discuss.tchncs.de on 2024-09-09 at 12:49

that ai accelerator will be used to spit out believably-looking but fucked up in subtle ways cobol. that’s job security with a side of eldritch horrors

=> More informations about this toot | More toots from skillissuer@discuss.tchncs.de

Written by froztbyte@awful.systems on 2024-09-09 at 12:53

haha I did try to ponder what the “LLMs, but cobol” corp world would look like but I couldn’t really think of anything particularly good, although this might be one

the other I thought of is:

I’m sure it’ll all go perfectly well

=> More informations about this toot | More toots from froztbyte@awful.systems

Written by BurgersMcSlopshot@awful.systems on 2024-09-10 at 19:00

I remember trying to get a not-stupid description of things like AI accelerators and NPUs, if I understand correctly isn’t it mostly “processes matrices of integers in less clock cycles?”

=> More informations about this toot | More toots from BurgersMcSlopshot@awful.systems