SoftestSapphic@lemmy.world to 196@lemmy.world · 5か月前"AI" rulelemmy.worldimagemessage-square89fedilinkarrow-up1693arrow-down154
arrow-up1639arrow-down1image"AI" rulelemmy.worldSoftestSapphic@lemmy.world to 196@lemmy.world · 5か月前message-square89fedilink
minus-squareBlaster M@lemmy.worldlinkfedilinkEnglisharrow-up23·5か月前So that assumes AI requests use 100 percent of the hardware 100 percent of the time.
minus-squareJohnDClay@sh.itjust.workslinkfedilinkarrow-up14arrow-down1·edit-25か月前Yes, but those servers are pretty ai specific, so that’s a decent assumption. Looks like Nvidia is drastically ramping up production of these servers, so current electricity use might be about 10x, I’m working on it.
minus-squareGissaMittJobb@lemmy.mllinkfedilinkarrow-up1·5か月前100% utilization 100% of the time? That seems like an unlikely figure, right?
minus-squareJohnDClay@sh.itjust.workslinkfedilinkarrow-up2·5か月前100% utilization yes, but server uptimes are in the 99 percent range.
minus-squareGissaMittJobb@lemmy.mllinkfedilinkarrow-up1·5か月前Idle consumption vs full utilization consumption are two very different things, though - this will have a large impact on the final number.
So that assumes AI requests use 100 percent of the hardware 100 percent of the time.
Yes, but those servers are pretty ai specific, so that’s a decent assumption. Looks like Nvidia is drastically ramping up production of these servers, so current electricity use might be about 10x, I’m working on it.
100% utilization 100% of the time? That seems like an unlikely figure, right?
100% utilization yes, but server uptimes are in the 99 percent range.
Idle consumption vs full utilization consumption are two very different things, though - this will have a large impact on the final number.