On 5 Jan 2000, Darrell Fuhriman wrote:
ndd -set /dev/tcp tcp_wscale_always 1 ndd -set /dev/tcp tcp_tstamp_always 1
Ah, interesting... I never considered whether DOT's stack supported the RFC1323 TCP extensions, but that might be part of the reason why my FreeBSD desktop (which turns on 1323 and 1644 by default) always seems to outrun much more expensive Sparc servers for network throughput, and with a lower CPU hit.
Note these are dedicated backup servers, so the very high tcp_X_hiwait may not be appropriate if you're not doing lots of bulk data transfer.
Do you find that with 2 x 400-MHz CPU's that you have enough cycles to stream 600 Mbps off the gigabit ethernet and sustain high throughput to your tape drives?