[ntp:questions] ntp server specs

Rick Jones rick.jones2 at hp.com
Thu Jun 16 20:39:44 UTC 2011


unruh <unruh at wormhole.physics.ubc.ca> wrote:
> On 2011-06-16, Rick Jones <rick.jones2 at hp.com> wrote:
> > unruh <unruh at wormhole.physics.ubc.ca> wrote:
> >> But terrible on 1000BaseT.
> >
> > I know it is a convenient shorthand to use the link-technology
> > name, but is it really that 1000BaseT as specified by the IEEE is
> > terrible, or the *implementation* specifics of various, perhaps
> > many network interface cards supporting 1000BaseT?  I'm thinking,
> > specifically of interrupt coalescing.

> No idea. All I know is that on my network before I put some 1000BT
> cards onto the network, the round trip time was very very stable at
> 130us. Now it flops around- sometimes 140 us, sometimes 280,
> sometimes 10ms. It (whatever it is) has destroyed a good timekeeping
> system.

Then sift through the bathwater man and see what has fouled it before
you toss-out the baby. :) And utter sweeping generalizations while
it flies through the air...

FWIW, here is some netperf output from my own little 1GbE network in
my cubicle:

raj at tardy:~/netperf2_trunk$ src/netperf -H 192.168.1.3 -t TCP_RR -l 10 -v 2
MIGRATED TCP REQUEST/RESPONSE TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.3 (192.168.1.3) port 0 AF_INET : histogram : demo
Local /Remote
Socket Size   Request  Resp.   Elapsed  Trans.
Send   Recv   Size     Size    Time     Rate         
bytes  Bytes  bytes    bytes   secs.    per sec   

16384  87380  1        1       10.00    12108.71   
16384  87380 
Alignment      Offset         RoundTrip  Trans    Throughput
Local  Remote  Local  Remote  Latency    Rate     10^6bits/s
Send   Recv    Send   Recv    usec/Tran  per sec  Outbound   Inbound
    8      0       0      0   82.585   12108.712 0.097     0.097 

Histogram of request/response times
UNIT_USEC     :    0:    0:    0:    0:    0:    0:    0:    0:    0:    0
TEN_USEC      :    0:    0:    0:    0:    0:  642:  523: 15767: 99307: 1053
HUNDRED_USEC  :    0: 3575:  180:   20:    5:    2:    3:    2:    0:    0
UNIT_MSEC     :    0:    5:    2:    1:    1:    1:    0:    0:    0:    0
TEN_MSEC      :    0:    0:    0:    0:    0:    0:    0:    0:    0:    0
HUNDRED_MSEC  :    0:    0:    0:    0:    0:    0:    0:    0:    0:    0
UNIT_SEC      :    0:    0:    0:    0:    0:    0:    0:    0:    0:    0
TEN_SEC       :    0:    0:    0:    0:    0:    0:    0:    0:    0:    0
>100_SECS: 0
HIST_TOTAL:      121089

one that I've not tried to tweak all that much, and where one of the
nodes remains connected to a reasonably busy "timeshare" network.

rick jones
-- 
I don't interest myself in "why." I think more often in terms of
"when," sometimes "where;" always "how much."  - Joubert
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...




More information about the questions mailing list