[Cryptech Core] capacity/performance numbers?

[Fredrik Thulin]

On onsdag 15 juni 2016 kl. 20:33:41 CEST Phil Roberts wrote:
> Understood on all points.
> 
> I think that for the folks who are interested in testing in "something
> like² an operational environment, a larger key store will be needed.  I
> don¹t think that entails a request for optimization of space.

We currently use flash "blocks" of the same size regardless of key type (4864 
bytes).

With the current flash chip, and with no effort at all to minimize the storage 
needed per key, my math says we should be able to store around 3000 RSA-8192 
keys or EC keys (of any size). This does require some coding though. We should 
be able to go from the current 6 to 12 by just changing some parameters.

With optimizations, we should be able to store maybe 6000 RSA-4096 keys, or 
maybe 40000 EC P-251 keys or maybe 80000 EC P-256 keys. 

This is the finger high up in the air, but it should be the right ballpark.

Hope that will give comfort to the testers.

/Fredrik