5 ms·

You are missing the Microsoft partnership. MS has a 48% stake in OpenAI and provides all the infrastructure through Azure, including purpose built machines for

You are missing the Microsoft partnership. MS has a 48% stake in OpenAI and provides all the infrastructure through Azure, including purpose built machines for model training. Microsoft has also launched GitHub Copilot, summarization features in Teams, and is widely reported to be adding GPT features to Bing.

scarmig 4y ago
Google has far more resources for training models and inference. Likely more than all their nearest competitors combined.
- redox99 4y ago
  Source?
  - dougmwne 4y ago
    There’s definitely no legitimate source for that info. Details of the exact machines that sit in the Azure and Google clouds are proprietary.
- yunwal 4y ago
  If you’re talking about physical hardware, Google pales in comparison to Amazon, and Gcloud is still smaller than Azure. It’s possible google’s private compute makes up for the Azure difference but it’s not like they’re in different leagues in terms of access to hardware.
  - alsodumb 4y ago
    Unlike Azure and Amazon, Google doesn't have to rely on Nvidia GPUs for training and inference, they achieved significant performance gains by using their custom TPUs.
    - coder543 4y ago
      Your comment is simply incorrect. Amazon has had their own TPU equivalents for training and inference for years: https://aws.amazon.com/machine-learning/trainium/ https://aws.amazon.com/machine-learning/trainium/ https://aws.amazon.com/machine-learning/inferentia/ https://aws.amazon.com/machine-learning/inferentia/ I really don't think this would be a limiting factor regardless, even if Amazon didn't already have multiple generations of these products. It's not as if an Amazon or Microsoft sized company is incapable of developing custom silicon to meet an objective, once an objective is identified. TPUs also aren't really that complicated to design, at least compared to GPUs. I'm slightly surprised Microsoft hasn't bothered to release any custom ML chips for Azure yet, but I guess they've run the numbers and decided to focus on other priorities for now.
  - scarmig 4y ago
    > google’s private compute
  - the-rc 4y ago
    Does it pale? How? AWS is very opaque when it comes to power demand, but you could use that to estimate machine counts and they admitted to Greenpeace that in 2016 they used 500-650MW — a wide range that obviously obfuscates on purpose. See the last Clicking Clean report from 2017. Google/Alphabet used 6514 terawatt hours in 2016 according to data collected from yearly reports at https://www.statista.com/statistics/788540/energy-consumption-of-google/ https://www.statista.com/statistics/788540/energy-consumptio... If my math is right, dividing that by 8760 hours in a year, you get 743MW. Of course, that also includes office space, etc. (did the AWS number, too?) but it should be clear that, cross checking with data center builds, optic fiber and energy purchases as well, for years Google+YT+Apps+GCP were larger than Amazon and all other AWS customers combined. I didn't even factor in efficiency, something that Amazon started focusing on quite a bit later. Someone might be able to extrapolate both numbers to today based on infrastructure, other metrics or other spend in quarterly financial statements (or power procurement, which will be complicated by the non trivial Amazon vs AWS distinction). All of the above to say that Amazon probably has more compute now, but it's a stretch to talk about "paling".
    - killjoywashere 4y ago
      Yeah. I have seen internal numbers on YouTube's daily ingest a few times over the years and every time my jaw drops. Like, the number from 2020 is ridiculous compared to their own number from 2017, and that was ridiculous.
      - the-rc 4y ago
        I have an idea what you're talking about, because I worked with the YT folks to reduce their Colossus costs and on Google storage in general until 2015. Another humbling and iluminating experience was comparing the Borg resources in the main crawling/indexing cluster to the Top500 list of the same period, something that always comes to mind and makes my eyes go wide when people compare DDG to Google. Or the day when a colleague taped a page saying only ”1E” next to our cubicle because we had just reached that much storage across the fleet.
  - flakeoil 4y ago
    But you are confusing yourself. Just because AWS has more servers being rented to and used by their customers doesn't necessarily mean they have the most CPU power available for themselves to run AI models. I also doubt computing power is the real bottleneck. Anyone of these companies (and most other companies too) can build enough large servers sites and they have the money. The costly and difficult part is the engineering resources, doing the right thing technically (AI wise) and business wise, not lose time, not bet on the wrong horse etc.
  - fomine3 4y ago
    I suspect that Google's own usage (search, YouTube, CDN, etc) is still bigger than GCP, is it correct?
paganel 4y ago
MS cannot deliver successful stuff on the web at scale. They can deliver unsuccessful stuff on the web at scale, they can deliver successful stuff that turns out to be inconsequential for their bottom line on the web in the long-term (AJAX came from MS), but it's just not in their DNA to take over the web. They had lots of chances to do it during the last 20 years or so, they had all the silver bullets at their disposal, they just couldn't deliver what it took.
- moufestaphio 4y ago
  Uh sorry what? Azure is the second largest cloud provider, well ahead of google.. That's like the definition of 'web scale' Not sure what decade you're stuck in here with comments about ajax
  - sidibe 4y ago
    Not agreeing with GP but the capacity of their public clouds is very different from total capacity. Amazon and Microsoft have bigger public clouds, but Google's own workload is probably bigger than anyone else's or even the public clouds by a large factor.
  - paganel 4y ago
    > Azure is the second largest cloud provider, well ahead of google. I explicitly mentioned the "web", as in, what we're doing right now on this website. Leaving aside the fact that Azure is mostly used big corporate/government entities, there's no web-startup that will potentially dominate the web and that would go into Azure just as.
    - rezonant 4y ago
      What makes you think anything being discussed in this thread is specific to HTML and web browsers?
- ctvo 4y ago
  > MS cannot deliver successful stuff on the web at scale. It's this type of confidence that you come back to HN over and over for. Let's ignore Azure, Office 365, and Microsoft's other online properties. In just video games: Microsoft owns and operates the Xbox network with over 100 million active monthly users[1]. That runs on the internet with all the features of a social network and more. I think they can deliver successful stuff at web scale OK, whatever their other shortcomings. 1 - https://hothardware.com/news/xbox-live-surpasses-100-million-active-monthly-users-gamepass https://hothardware.com/news/xbox-live-surpasses-100-million...
mountainriver 4y ago
Yeah they also trash up the brand. Seeing ChatGPT be thrown into Teams and Bing let’s me know where it’s headed

scarmig 4y ago

Google has far more resources for training models and inference. Likely more than all their nearest competitors combined.

redox99 4y ago
Source?
- dougmwne 4y ago
  There’s definitely no legitimate source for that info. Details of the exact machines that sit in the Azure and Google clouds are proprietary.
yunwal 4y ago
If you’re talking about physical hardware, Google pales in comparison to Amazon, and Gcloud is still smaller than Azure. It’s possible google’s private compute makes up for the Azure difference but it’s not like they’re in different leagues in terms of access to hardware.
- alsodumb 4y ago
  Unlike Azure and Amazon, Google doesn't have to rely on Nvidia GPUs for training and inference, they achieved significant performance gains by using their custom TPUs.
  - coder543 4y ago
    Your comment is simply incorrect. Amazon has had their own TPU equivalents for training and inference for years: https://aws.amazon.com/machine-learning/trainium/ https://aws.amazon.com/machine-learning/trainium/ https://aws.amazon.com/machine-learning/inferentia/ https://aws.amazon.com/machine-learning/inferentia/ I really don't think this would be a limiting factor regardless, even if Amazon didn't already have multiple generations of these products. It's not as if an Amazon or Microsoft sized company is incapable of developing custom silicon to meet an objective, once an objective is identified. TPUs also aren't really that complicated to design, at least compared to GPUs. I'm slightly surprised Microsoft hasn't bothered to release any custom ML chips for Azure yet, but I guess they've run the numbers and decided to focus on other priorities for now.
- scarmig 4y ago
  > google’s private compute
- the-rc 4y ago
  Does it pale? How? AWS is very opaque when it comes to power demand, but you could use that to estimate machine counts and they admitted to Greenpeace that in 2016 they used 500-650MW — a wide range that obviously obfuscates on purpose. See the last Clicking Clean report from 2017. Google/Alphabet used 6514 terawatt hours in 2016 according to data collected from yearly reports at https://www.statista.com/statistics/788540/energy-consumption-of-google/ https://www.statista.com/statistics/788540/energy-consumptio... If my math is right, dividing that by 8760 hours in a year, you get 743MW. Of course, that also includes office space, etc. (did the AWS number, too?) but it should be clear that, cross checking with data center builds, optic fiber and energy purchases as well, for years Google+YT+Apps+GCP were larger than Amazon and all other AWS customers combined. I didn't even factor in efficiency, something that Amazon started focusing on quite a bit later. Someone might be able to extrapolate both numbers to today based on infrastructure, other metrics or other spend in quarterly financial statements (or power procurement, which will be complicated by the non trivial Amazon vs AWS distinction). All of the above to say that Amazon probably has more compute now, but it's a stretch to talk about "paling".
  - killjoywashere 4y ago
    Yeah. I have seen internal numbers on YouTube's daily ingest a few times over the years and every time my jaw drops. Like, the number from 2020 is ridiculous compared to their own number from 2017, and that was ridiculous.
    - the-rc 4y ago
      I have an idea what you're talking about, because I worked with the YT folks to reduce their Colossus costs and on Google storage in general until 2015. Another humbling and iluminating experience was comparing the Borg resources in the main crawling/indexing cluster to the Top500 list of the same period, something that always comes to mind and makes my eyes go wide when people compare DDG to Google. Or the day when a colleague taped a page saying only ”1E” next to our cubicle because we had just reached that much storage across the fleet.
- flakeoil 4y ago
  But you are confusing yourself. Just because AWS has more servers being rented to and used by their customers doesn't necessarily mean they have the most CPU power available for themselves to run AI models. I also doubt computing power is the real bottleneck. Anyone of these companies (and most other companies too) can build enough large servers sites and they have the money. The costly and difficult part is the engineering resources, doing the right thing technically (AI wise) and business wise, not lose time, not bet on the wrong horse etc.
- fomine3 4y ago
  I suspect that Google's own usage (search, YouTube, CDN, etc) is still bigger than GCP, is it correct?

paganel 4y ago

MS cannot deliver successful stuff on the web at scale. They can deliver unsuccessful stuff on the web at scale, they can deliver successful stuff that turns out to be inconsequential for their bottom line on the web in the long-term (AJAX came from MS), but it's just not in their DNA to take over the web. They had lots of chances to do it during the last 20 years or so, they had all the silver bullets at their disposal, they just couldn't deliver what it took.

moufestaphio 4y ago
Uh sorry what? Azure is the second largest cloud provider, well ahead of google.. That's like the definition of 'web scale' Not sure what decade you're stuck in here with comments about ajax
- sidibe 4y ago
  Not agreeing with GP but the capacity of their public clouds is very different from total capacity. Amazon and Microsoft have bigger public clouds, but Google's own workload is probably bigger than anyone else's or even the public clouds by a large factor.
- paganel 4y ago
  > Azure is the second largest cloud provider, well ahead of google. I explicitly mentioned the "web", as in, what we're doing right now on this website. Leaving aside the fact that Azure is mostly used big corporate/government entities, there's no web-startup that will potentially dominate the web and that would go into Azure just as.
  - rezonant 4y ago
    What makes you think anything being discussed in this thread is specific to HTML and web browsers?
ctvo 4y ago
> MS cannot deliver successful stuff on the web at scale. It's this type of confidence that you come back to HN over and over for. Let's ignore Azure, Office 365, and Microsoft's other online properties. In just video games: Microsoft owns and operates the Xbox network with over 100 million active monthly users[1]. That runs on the internet with all the features of a social network and more. I think they can deliver successful stuff at web scale OK, whatever their other shortcomings. 1 - https://hothardware.com/news/xbox-live-surpasses-100-million-active-monthly-users-gamepass https://hothardware.com/news/xbox-live-surpasses-100-million...

mountainriver 4y ago

Yeah they also trash up the brand. Seeing ChatGPT be thrown into Teams and Bing let’s me know where it’s headed