The US attempt to slow down China's technological development succeeds on the basis of preventing China from directly following the same path, but may backfire in the sense it forces innovation by China in a different direction. The overall outcome for us all may be increase efficiency as a result of this forced innovation, especially if Chinese companies continue to open source their advances, so we may in the end h…
Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
241–250 of 331 posts
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#242Alibaba Cloud claims to reduce Nvidia GPU used for serving unpopular models by 82% (emphasis mine) > 17.7 per cent of GPUs allocated to serve only 1.35 per cent of requests in Alibaba Cloud’s marketplace, the researchers found Instead of 1192 GPUs they now use 213 for serving those requests.
14.5% is worth a raise at least. But it’s still misleading.
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#243Earlier quoted context omitted.
Models take a lot of VRAM which is tightly coupled to the GPU so yeah, it's basically sitting there with the model waiting for use. I'm sure they probably do idle out but a few minutes of idle time is a lot of waste--possibly the full 82% mentioned. In this case they optimized by letting the GPUs load multiple models and sharing the load out by token.
They definitely won't idle out- if they idle out, it'll take on the order of up to 60 seconds to load the model back into VRAM, depending on the model. That's an eternity for a request. I highly doubt they will timeout any model they serve.
I'd have to play with the configuration and load calcs, but I'm sure there's a low param, neat solution to the request/service problem.
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#244Earlier quoted context omitted.
This statement doesn't make a lot of sense. 40-50% of vehicles are foreign-made already[1]. I would strongly wager it's vastly more likely that these Chinese vehicles do not meet US safety standards - which are quite high. [1] https://www.whitehouse.gov/fact-sheets/2025/03/fact-sheet-pr...
Much higher standards than European ones? Because the cars do sell there as well.
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#245Earlier quoted context omitted.
In the past, software and computer engineers would tackle problems head-on, designing algorithms and finding creative solutions. thanks to the US restrictions on semiconductor industry (Chinese), Chinese engineers are being forced to innovate and find their own ways to overcome challenges like the old school engineers (What Silicon Valley used to be)
If you're one who sees progress as an end goal unto itself, what you describe is a good thing. When one party is attempting novel solutions to outcompete the competition we will be faster to whatever the next change is. That said, I'm not sure what the US policies specifically have to do with this. Countries are always in competition with one another, and if one industry or technology is considered a national securit…
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#246The US attempt to slow down China's technological development succeeds on the basis of preventing China from directly following the same path, but may backfire in the sense it forces innovation by China in a different direction. The overall outcome for us all may be increase efficiency as a result of this forced innovation, especially if Chinese companies continue to open source their advances, so we may in the end h…
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#247Earlier quoted context omitted.
Models take a lot of VRAM which is tightly coupled to the GPU so yeah, it's basically sitting there with the model waiting for use. I'm sure they probably do idle out but a few minutes of idle time is a lot of waste--possibly the full 82% mentioned. In this case they optimized by letting the GPUs load multiple models and sharing the load out by token.
They definitely won't idle out- if they idle out, it'll take on the order of up to 60 seconds to load the model back into VRAM, depending on the model. That's an eternity for a request. I highly doubt they will timeout any model they serve.
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#248Earlier quoted context omitted.
I do. I don't know what gave people the impression that I was born an American and went to their schools. In "open societies" like the US, the form information control are rather sophisticated, but they are there. However, I grew up in a one TV channel country. I learned to never trust the news. This immunizes me from state propaganda a little bit. I think I was more referring to general abundance of credulity that i…
> I don't know what gave people the impression that I was born an American and went to their schools. The fact that your comment appears to defend the US position, by virtue of attacking a press release from a major non-US competitor by questioning its source. > This story is just as ridiculous. AliBaba claims to reduce usage of equipment from a business that is being banned by their government. Magically. And during…
It only appears that way because you see the world through a simplistic bipolar lens.
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#249The US attempt to slow down China's technological development succeeds on the basis of preventing China from directly following the same path, but may backfire in the sense it forces innovation by China in a different direction. The overall outcome for us all may be increase efficiency as a result of this forced innovation, especially if Chinese companies continue to open source their advances, so we may in the end h…
I think anti-immigrant rhetoric will have the most impact against the US. A lot of the people innovating on this stuff are being maligned and leaving in droves. Aside from geography, attracting talent from all over the world is the one edge the US has a nation over countries like China. But now the US is trying to be xenophobic like China, restrict tech import/export like China but compete against 10x population and…
Re: Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system
#250Earlier quoted context omitted.
History has shown that withholding technology from China does not significantly stop them and they'll achieve it (or better) in a small number of years. In many senses there's hubris in the western* view of China accomplishments: most of what western companies have created has had significant contribution by Chinese scientists or manufacturing, without which those companies would have nothing. If you look at the name…
The whole “China copies everything” narrative is becoming less and less true. It’s funny - it’s at the point with Chinese manufacturing for niche electronic goods (e.g rooftop van air conditioner) where some Chinese brands are more trustworthy - more value for your money and sometimes even better overall quality. With American brands you gotta make sure you’re not overpaying for dated tech that is inefficient. Maybe…
Enterprises often prefer having US based support and so can prefer US or European machines that have that supply chain setup.