> such a cloud would be useless for this application
Either you don't know what is in that DC, or you do. If you don't know what's in the DC, you can't say one way or the other about what it can and can't do. If you do know what's in the DC, then you'd likely be prevented from saying what it's capabilities were or would deny those capabilities publicly if it did possess them. This is why people make tinfoil hats; because they reach cognitive dissonance quickly with logic built around untrusted data.
Because I can't implicitly trust you, or Amazon, or the intelligence agencies, I'm left with the reasoning capabilities I have and trust. Floundering in cognitive dissonance isn't going to get me anywhere in that regard. I'm left with what I know, which is that US agencies pay Amazon a staggering amount of money to run a datacenter for them. I know that AWS itself runs servers with special purpose hardware in them, such as GPUs, and that most of the value in AWS comes from federating systems and providing fault tolerance and easy programmatic access to provisioning. What sits behind all those features is anyone's guess. Amazon does not publicly discuss their relationship with the government, so that's zero help here.
If Amazon wanted to own the world's public cloud AND remain trusted, they should have thought twice about doing a deal with the government. In Germany, this shit wouldn't fly. Companies who provide services to the government there sign special agreements to ensure what services they provide to the government aren't colluded with the services they provide to individuals. It's beyond dumb that we allow this in the US and that the largest cloud provider here is mute in that regard.
"If you don't know what's in the DC, you can't say one way or the other about what it can and can't do."
Sure you can: just go with worst case scenario and assume custom hardware. I did that on Schneier's blog below:
"The room for error here is in the ASIC assessment. We need to figure out how much they can parallelize this, either in cores or custom circuits, in a given ASIC. Then, how many of those can do in one chip at 14-28nm. Then how many they can squeeze in a rack. Then how much factoring can be done with Amazon or Microsoft datacenters full of those. That would be the most paranoid assessment.
You see, the unit prices of these things are really low vs initial development costs. That's one of main drivers for hardware to shrink. The chips might cost pro's $10-30 million to develop with boards a tiny fraction of that. Then, the chips themselves are fabbed dirt-cheap with the boards being inexpensive. If algorithm doesn't need much communication, then they can use standard I/O options to farm out the jobs which then just run until they complete. They could add more capacity year after year cheaply with incremental energy cost after spending a ton of money on chip design and real estate just once. They could get 50,000-100,000 chips with multiple accelerators on each every year.
So, I think the authors upper bound is lower than the real upper bound. Need to get specialists who have implemented algorithms like this in hardware to show how it will likely be implemented. Need at least one person whose done a 28nm design to estimate how much of the chip can be dedicated to that with the other functions considered. Then multiply that by whatever Amazon has to get a decent upper-bound. "
Either you don't know what is in that DC, or you do. If you don't know what's in the DC, you can't say one way or the other about what it can and can't do. If you do know what's in the DC, then you'd likely be prevented from saying what it's capabilities were or would deny those capabilities publicly if it did possess them. This is why people make tinfoil hats; because they reach cognitive dissonance quickly with logic built around untrusted data.
Because I can't implicitly trust you, or Amazon, or the intelligence agencies, I'm left with the reasoning capabilities I have and trust. Floundering in cognitive dissonance isn't going to get me anywhere in that regard. I'm left with what I know, which is that US agencies pay Amazon a staggering amount of money to run a datacenter for them. I know that AWS itself runs servers with special purpose hardware in them, such as GPUs, and that most of the value in AWS comes from federating systems and providing fault tolerance and easy programmatic access to provisioning. What sits behind all those features is anyone's guess. Amazon does not publicly discuss their relationship with the government, so that's zero help here.
If Amazon wanted to own the world's public cloud AND remain trusted, they should have thought twice about doing a deal with the government. In Germany, this shit wouldn't fly. Companies who provide services to the government there sign special agreements to ensure what services they provide to the government aren't colluded with the services they provide to individuals. It's beyond dumb that we allow this in the US and that the largest cloud provider here is mute in that regard.