4 ms·
I know we won't get it this from FB, but I'd be really interested to see how the relationship of compute power to engineering hours scales. They mention custom
by islewis 3y ago
I know we won't get it this from FB, but I'd be really interested to see how the relationship of compute power to engineering hours scales.
They mention custom building as much as they can. If FB magically has the option to 10x the compute power, would they need to re-engineer the whole stack? What about 100x? Is each of these re-writes just a re-write, or is it a whole order of magnitude more complex?
My technical understanding of what's under the hood of these clusters is pretty surface level- super curious if anyone with relevant experience has thoughts?
- bilekas 3y agoI'm not 100% sure but I would.make an educated guess that that cluster in the first image for example is a sample of scalable clusters, so throwing more hardware at it could bring improvements but sooner or later the cost to improvements will call for an optimization or rewrite as you call it, so a bit of both usually. It seems a bit of a balancing act really!
- tintor 3y ago"just a re-write"
- mirekrusin 3y ago...the idea is that at some point it "just re-writes" itself.
- ametrau 3y agoThe day after that, we have true AGI.
- jvalencia 3y agoThe cost of training quickly outpaces the cost of development as context length increases. So hardware is cheap until it isn't anymore, by orders of magnitude.
- samstave 3y agoBut there is still significant cost in the physical buildouts of new pods/DCs, whatever and the human engineering hours to physically build, even though its a mix of resources across the vendors and FB? - it still would be interesting to know man hours into the physical build of the HW.