Hi all,following Apple's M1 CPU announcement, I have been wondering, how one could best utilize something like that? (purely hypothetical and ignoring, that there are no wide-spread ARM server CPUs somewhat with the M1 profile as well as arch support etc. pp.)
I.e., how could one schedule such a heterogeneous CPU with a big.little-like design, i.e., separate(?) the 'big cores' from the 'little cores' - assuming that with HTC the whole CPU w/should be utilized ~100% for most of the time? Binding jobs with cpuset to specific cores (no idea, if it is possible wih ARMs?) and have a node offer each core class as different slots?
Cheers, Thomas
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature