> cgo is not go.
The speed of system calls is similar though -- 110 microseconds for
a UDP send/receive pair over loopback. As compared to a single core of
a 2.4GHz laptop, the MIPS nodes are:
- 10 to 20 times slower on integer code;
- 30 times slower on small UDP packets over loopback;
- 30 times slower on Cgo calls;
- almost 1000 times slower on floating point (softfloat).
I haven't measured TCP throughput yet.
While I find the benchmarks somewhat disappointing, what that means is
that the MIPS nodes are more than fast enough for my use case (some
simple network monitoring), so many thanks to whoever did the MIPS32 and
softfloat support. Or perhaps not, now I no longer have any excuse to
buy myself a bunch of Beaglebones.
While I have everyone's ear -- does anyone know why 1.10 doesn't attempt
to fill the delay slots? The code is there, but it's disabled.
--
You received this message because you are subscribed to the Google Groups
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
