You cant fit a Stratix 10 on a nVME stick... Ive tried. Kintex or Arrria is about as big as you can get. Damn 22x80 form factor.
I know.
I question the need for ultra large and expensive FPGAs instead of clusters of smaller FPGAs with high speed interconnects.
Perhaps a custom PCI-E format board with 4-6 last generation FPGA devices with some fast memory and a cross point switch. Are there many individual components of the Xnn series of algos which won't fit on an arria or even a cyclone ?
Thats exactly what Ive been working on, for reasonable definitions of small and fast.
I have two active projects in the first spin batch phase. One is nVME with basically the biggest thing you can fit on there, and it augments GPUs more than works standalone.
The second is 4 chips on one PCIe card, with a switch, but the most reasonable chip that can be used in that configuration is still not what you would call cheap. Frankly even the nVME chip is as much as some graphics cards to get 4x 3.0 PCIe lanes.
The one advantage is the 4-chip board uses modules, so you could buy one with 1 module populated in the 3 figure range. When it is ready, which is likely August at this point for mass production.