|
Posted by Walter Roberson on October 2, 2006, 3:05 pm
If you were Registered and logged in, you could reply and use other advanced thread options
[snipping the attribution for my words... tsk tsk]
>> How far did you go in checking for a duplex mismatch?
>>>The workstations are Apple G5s and the server is a Dell PowerEdge 1650.
>> Is that PowerEdge running Windows 2000 or Windows XP? I have seen
>> Windows 2000 -lie- about the duplex that it negotiated: every level
>> on the W2K server reported that the link was full duplex,
>Without a Fluke Lanmeter, is there any way to know 100% for sure what the
>Win2K machine is negotiating?
Try the other approach: nail the interface speeds and see what happens.
However, you mentioned that this is gigabit ports; autonegotiation
of some parameters is mandatory for gigabit (that is actually being
used as gigabit.) The extent to which you can lock down the ports
to particular speeds or duplexes might be limited.
Let's see... if you were to lock the ports to 100/Full then you'd
have 12.5 MB/s of bandwidth going in to the PowerEdge, share that
equally between two systems and the average would be about 6 MB/s each,
which by coincidence is the same figure you mention achieving now.
It'd be an interesting test, to lock both to 100/Full and see if you
get the 6 MB/s each...
|
|
Posted by DLR on October 2, 2006, 3:26 pm
If you were Registered and logged in, you could reply and use other advanced thread options
Chris McFarling wrote:
>> Sounds like a duplex mismatch.
>
> One of the first things I checked. There is no duplex mismatch.
>
>> "gigabit NICs" and "gigabit switch" doesn't tell us anything about the
>> real throughput capacities of the PCs or of the switch. You can put
>> a gigabit interface onto a tin-can telephone, if you have a way of
>> dealing with the buffering.
>
> The workstations are Apple G5s and the server is a Dell PowerEdge 1650. All
> subsystems for each computer have been checked and are not the limiting
> facter here.
>
>> Overflowing a switch's buffers is possible but when you get that big
>> of a difference in speeds, the reason is almost always duplex mismatches.
>
> The switch is an HP 4108gl. Is this performance pattern consistent with a
> buffer overload scenario? I believe the buffers on this switch are pretty
> small, like 1MB.
>
I think that is a managed switch. If so you can telnet into it and see
what the error counts look like. Then check out the various other
details and stats on the ports. Plus you can see what the switch is
connecting at and the G5s are doing. On the G5s look under the last tab
on the network control panel for the Ethernet port on the Network System
Preference.
|
|
Posted by Rick Jones on October 2, 2006, 2:21 pm
If you were Registered and logged in, you could reply and use other advanced thread options
>>This is a general switch architecture question. Let's say there are
>>2 workstations communicating with a server. All three nodes are
>>using gigabit NICs and they're connected to a gigabit
>>switch. Workstation A can send data to the server at 35MB/s when
>>only workstation A is sending. Likewise workstation B can send data
>>to the server at 35MB/s when only workstation B is sending. When
>>both workstation A & workstation B send data to the server
>>simultaneously however, the transfer rate for both drops to about
>>6MB/s.
> Sounds like a duplex mismatch.
?!? Duplex mismatch would have been visible with just one client
sending to the server. It wouldn't require both sending at once.
Having said that, it _would_ be good for Chris to check statistics
everywhere possible - TCP level on clients and servers (eg netstat)
link-level statistics on clients and servers (OS specific) and swtich
stats.
> Overflowing a switch's buffers is possible but when you get that big
> of a difference in speeds, the reason is almost always duplex
> mismatches.
I would say packet losses, _possibly_ from duplex, but could be other
things.
rick jones
--
oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag
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...
|
|
Posted by Walter Roberson on October 2, 2006, 2:40 pm
If you were Registered and logged in, you could reply and use other advanced thread options
>>>using gigabit NICs and they're connected to a gigabit
>>>switch. Workstation A can send data to the server at 35MB/s when
>>>only workstation A is sending. Likewise workstation B can send data
>>>to the server at 35MB/s when only workstation B is sending. When
>>>both workstation A & workstation B send data to the server
>>>simultaneously however, the transfer rate for both drops to about
>>>6MB/s.
>> Sounds like a duplex mismatch.
>?!? Duplex mismatch would have been visible with just one client
>sending to the server. It wouldn't require both sending at once.
35 MB/s is only 280 Mb/s. We can assume there is no great latency
in the circuit as no WAN or long distance link was mentioned, just
a single gigabit switch. In a follow-up the poster said Mac G5;
I wouldn't think those would have any problem transmitting more than
280 Mb/s. The Dell PowerEdge 1850 I'd have to investigate the
architecture of.
Anyhow, there is the possibility that there is -already- some duplex
problem just with a single client; the effect is possibly being reduced
with SAK.
When both clients are going, there are more ACKs coming back, and more
duplex clash opportunities. -Maybe- even a reduced window size
relative to the single client.
Mind you, I would ask about the quality of the gigabit switch,
the number of ports, and which device is plugged into which port.
If it has more than 5 ports, it probably has more than one gigabit
chip in it, and greater throughput might -potentially- be achievable
by redistributing the connections so that the big transmitters
(the clients) are on differerent chips (thus different buffers.)
|
|
Posted by Rick Jones on October 2, 2006, 3:02 pm
If you were Registered and logged in, you could reply and use other advanced thread options
>>> Sounds like a duplex mismatch.
>>?!? Duplex mismatch would have been visible with just one client
>>sending to the server. It wouldn't require both sending at once.
> 35 MB/s is only 280 Mb/s. We can assume there is no great latency
> in the circuit as no WAN or long distance link was mentioned, just
> a single gigabit switch. In a follow-up the poster said Mac G5;
> I wouldn't think those would have any problem transmitting more than
> 280 Mb/s.
Modulo the default socket buffer size(s) under Mac OS X... To take
file systems out of the equation it might be good for the OP to run
some netperf and play around with socket buffer/window sizes.
rick jones
--
denial, anger, bargaining, depression, acceptance, rebirth...
where do you want to be today?
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...
|
| Similar Threads | Posted | | Switch to switch causing traffic problems? | April 19, 2005, 11:46 am |
| difference b/w inter-switch link port and trunk port | July 6, 2008, 12:06 pm |
| Port on switch going crazy........ | October 2, 2006, 11:48 am |
| 16 port unmanaged switch | April 15, 2007, 4:22 pm |
| Sharing a switch port | August 30, 2008, 3:17 pm |
| Switch Killed Computer's LAN port | April 2, 2005, 11:04 pm |
| what gigabit switch (8 port) to select | March 6, 2007, 1:23 pm |
| Assign static IPs to port of a managed switch | June 24, 2005, 10:42 am |
| Media converter vs Switch with Fiber port | January 12, 2006, 1:04 am |
| obtaining port bits per second data from switch | August 29, 2006, 12:20 pm |
|
|
Yahoo! 
Google 
Windows Live 
del.icio.us 
digg 
Netscape 
XML site map