So why *aren't* there small bits of compute baked into RAM yet? Referring to highly local operations with fixed dimensionality, or reducing queries (return or update memory region where following constraint is true). We have TB+ EC2 nodes, after all. Why move the data?
-
-
The more you think you know...
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
All that is still moving the data, with ever larger busses. I'm wondering if we can get away with not actually moving it, something like making DRAM refresh "smart" or "streaming".
-
The thing is DRAM doesn't spend a huge fraction of the time refreshing itself. DRAM is pipelined and you can stream data in/out of RAM pretty much as fast as the physical technology allows.
- Show replies
New conversation -
-
-
oh, TIL. I thought eDRAM was always different-die.
-
Yeah, half the Wii U's GPU/SoC die is eDRAM. Two blocks of different density. I think the top left bit is true SRAM. Both that & the smaller eDRAM block used to be eDRAM on Wii/GC, which was also half the die. https://marcan.st/transf/latte_stitched.jpg … https://marcan.st/transf/hollywood_die.jpg … (warning: huge).pic.twitter.com/Mn4FaVn1rL
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.