I have ~120 Ekeys/s in BSGS/keyhunt on AMD after used the AOCC compiler

Is this the speed at which public keys are checked in the hash table (using a bloom filter?), or is this the real speed at which the processor generates public keys?
Quite a high speed even for BSGS on a video card
120 exakeys = 120.000.000.000 gigakeys
Pretending some 4Ghz CPU generates, sequentially, one key per cycle (it doesn't, more like one key over 300 cycles on average, and that's with all possible optimizations) it would still take 30.000.000.000 CPU cores to reach that speed.
Or (300 cycles/key): 9.000.000.000.000 cores that are running all at 100% with no OS, nothing else running, all working at full speed doing nothing except crunching numbers inside the CPU registers.
I think maybe that speed reflects space coverage rather than operating time, and space coverage speed is logarithmic not linear.