By: Ireland (boh.delete@this.outlook.ie), January 31, 2017 7:58 pm
Room: Moderated Discussions
tarlinian (tarlinian.delete@this.gmail.com) on January 31, 2017 6:38 pm wrote:
> Ireland (boh.delete@this.outlook.ie) on January 31, 2017 5:11 pm wrote:
>
> Am I the only one who finds this to be completely incoherent? Are you really claiming
> that the point of all this is to put a supercomputer on a boat in order to somehow
> improved weather predictions over standard weather forecasting models?
What exactly is wrong with that idea? We've got things such as micro-nuclear technology now, that we need to use - and dumping excess heat into the ocean - is easier than dumping it on land. Even though, locations such as Britain or Ireland are often preferred for the likes of data center construction. The best place for a supercomputer is out in the oceans surely? The best place for 'big data' is on dry land. And the best thing to connect one to the other, are the satellites for a means of communication. They're just not able to do what they need to do, in terms of forecasting longer term. It's one approach, and it could be made to function. I explained the basic outline of the architecture, earlier in this month. You can refer to it here, if you'd like to.
Business Intelligence
http://www.realworldtech.com/forum/?threadid=164183&curpostid=164183
The basic problems with things like 'Deepwater Horizon', was the fact that one had 'business intelligence' kinds of conversations, happening in places like the 'drilling shack', which should really only be places, where 'engineering model' kinds of conversations need to happen. I.e. All of that analysis, to do with time, money, risk etc is best separated from the core engineering model, that is made as small, and interactively coupled to the 'hardware' problem, or surveillance problem, as you can. I.e. The computing power, closest to where the weather happens first. That's not on the mainland, that's miles and miles out beyond, in those really nasty external and marine environments.
> Ireland (boh.delete@this.outlook.ie) on January 31, 2017 5:11 pm wrote:
>
> Am I the only one who finds this to be completely incoherent? Are you really claiming
> that the point of all this is to put a supercomputer on a boat in order to somehow
> improved weather predictions over standard weather forecasting models?
What exactly is wrong with that idea? We've got things such as micro-nuclear technology now, that we need to use - and dumping excess heat into the ocean - is easier than dumping it on land. Even though, locations such as Britain or Ireland are often preferred for the likes of data center construction. The best place for a supercomputer is out in the oceans surely? The best place for 'big data' is on dry land. And the best thing to connect one to the other, are the satellites for a means of communication. They're just not able to do what they need to do, in terms of forecasting longer term. It's one approach, and it could be made to function. I explained the basic outline of the architecture, earlier in this month. You can refer to it here, if you'd like to.
Business Intelligence
http://www.realworldtech.com/forum/?threadid=164183&curpostid=164183
The basic problems with things like 'Deepwater Horizon', was the fact that one had 'business intelligence' kinds of conversations, happening in places like the 'drilling shack', which should really only be places, where 'engineering model' kinds of conversations need to happen. I.e. All of that analysis, to do with time, money, risk etc is best separated from the core engineering model, that is made as small, and interactively coupled to the 'hardware' problem, or surveillance problem, as you can. I.e. The computing power, closest to where the weather happens first. That's not on the mainland, that's miles and miles out beyond, in those really nasty external and marine environments.