> > also, try to test the speed of the true random number generator > > (I don't know how to access it; maybe just dd /dev/urandom?) > > Here's one: > > ~% time dd if=/dev/urandom bs=1024 count=10240 of=/dev/null > 10240+0 records in > 10240+0 records out > 10485760 bytes transferred in 15.957354 secs (657111 bytes/sec) > dd if=/dev/urandom bs=1024 count=10240 of=/dev/null 0.00s user > 15.85s system 99% cpu 15.970 total > > ~% time dd if=/dev/random bs=1024 count=10240 of=/dev/null > 10240+0 records in > 10240+0 records out > 10485760 bytes transferred in 15.967514 secs (656693 bytes/sec) > dd if=/dev/random bs=1024 count=10240 of=/dev/null 0.01s user 15.85s > system 99% cpu 15.976 total >
I'm also going to be doing a lot of SSL on a Poweredge 2950. So the above is purely a memory/cpu test correct? (not like doing IO tests where you want the size of the test data to be at least twice physical ram). Here's results for the Woodcrest- I assume this is only on a single core. Yes, I realize that the woodcrest is faster than above mentioned cpu, but a 100x speed difference? That doesn't seem realistic to me (although if those are valid results, I'd be pretty happy with that)... bash-2.05b$ time dd if=/dev/urandom bs=1024 count=10240 of=/dev/null 10240+0 records in 10240+0 records out 10485760 bytes transferred in 0.154649 secs (67803598 bytes/sec) bash-2.05b$ time dd if=/dev/urandom bs=8k count=10240 of=/dev/null 10240+0 records in 10240+0 records out 83886080 bytes transferred in 1.165706 secs (71961615 bytes/sec) Thanks, Bucky _______________________________________________ [email protected] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hardware To unsubscribe, send any mail to "[EMAIL PROTECTED]"
