Evaluating STP Quality for Gigabit Ethernet
Bookmark this page:
 Yahoo!
 Google
 Windows Live
 del.icio.us
 digg
 Netscape
|
|
|||||||||||||||||||
|
Posted by Will on July 25, 2006, 2:35 am
Please log in for more thread options >
> Why do you believe that links are being run at 100, let > alone gig? Some very smart equipment falls back to 10. > Almost all will fall back to 100. Because the link lights on the switch, as well as the software application, both say gigabit speed for all attached links. -- Will | |||||||||||||||||||
|
Posted by Robert Redelmeier on July 25, 2006, 2:02 pm
Please log in for more thread options > Because the link lights on the switch, as well as the
> software application, both say gigabit speed for all > attached links. And do you actually see 13+ MByte/s transfer speeds? The lights are often wrong and the software may be if it's just talking to the etherchip. -- Robert | |||||||||||||||||||
|
Posted by Will on July 25, 2006, 5:21 pm
Please log in for more thread options
> And do you actually see 13+ MByte/s transfer speeds?
> The lights are often wrong and the software may be > if it's just talking to the etherchip. Yes, we seem to be getting about 12 to 20 MB/sec. I'm a little perplexed why we see the bandwidth peek around 20 MB/sec then go to zero for a while before jumping again to a peak. -- Will | |||||||||||||||||||
|
Posted by DLR on July 25, 2006, 6:24 pm
Please log in for more thread options
Will wrote: >> And do you actually see 13+ MByte/s transfer speeds?
>> The lights are often wrong and the software may be >> if it's just talking to the etherchip. >
We have a printer which does not correctly negotiate when connected to a
> Yes, we seem to be getting about 12 to 20 MB/sec. > > I'm a little perplexed why we see the bandwidth peek around 20 MB/sec then > go to zero for a while before jumping again to a peak. > gig link. We're waiting on a fix from the manufacturer. But until I looked at the error rates, it seemed to be a protocol problem. It would work fine if we sent jobs via AppleTalk. Say 1 minute before printing started. But if we used IP as the transport the same job would take 1 to 3 hours before it would print. When I checked the switch stats, CRC errors outnumbered good packets when doing IP. But until I looked at it everyone was saying our computers couldn't talk IP correctly as it did work with AppleTalk. Turned out the issue was the interface on the printer was much more likely to mishandle large packets when connected to a gig link. My point is once you have a link with issues, problems can manifest themselves in strange ways. | |||||||||||||||||||
|
Posted by Robert Redelmeier on July 25, 2006, 7:59 pm
Please log in for more thread options
> Yes, we seem to be getting about 12 to 20 MB/sec.
Definitely gig. > I'm a little perplexed why we see the bandwidth peek around 20
> MB/sec then go to zero for a while before jumping again to a peak. There are a number of possible explanations: If it's quick cycling (ms) it might just be track seek. If it's much slower (s) and erratic, then that may be all the traffic to be sent! -- Robert | |||||||||||||||||||
| Similar Threads | Posted |
| Evaluating STP Quality for Gigabit Ethernet | July 22, 2006, 9:15 pm |
| Cat 5e vs 6 and gigabit ethernet | February 4, 2006, 1:12 am |
| Gigabit Ethernet Cabling | November 17, 2004, 3:55 pm |
| Shielded Cat5e for Gigabit Ethernet | September 14, 2004, 9:35 pm |
| Patch Cable Quality - Brand Differences? | August 21, 2006, 10:28 am |
| Ok this may be a beaten up subject but...c5e vs c6 gigabit. | June 1, 2006, 10:11 am |
| Gigabit Cabling and Auto-MDIX Problems | May 24, 2005, 1:41 am |
| DIY Ethernet | September 10, 2004, 8:57 pm |
| Ethernet cabling used outside | July 29, 2004, 7:57 pm |
| Cat 5 for phone and ethernet? | December 20, 2004, 9:57 pm |
| Cat 5 for phone and ethernet? | December 20, 2004, 9:56 pm |
| PoE and Ethernet switches | October 1, 2005, 10:22 am |
| Re: ethernet polarity | November 4, 2005, 10:44 pm |
| Re: ethernet and AC in same conduit | January 2, 2009, 8:47 pm |
| Ethernet cable shield | November 23, 2004, 5:02 pm |
> > individual rooms uplinked to GSM7224 layer 2 switches in the
> > central room, and I have yet to see a single error reported
> > on any switch. I'm not believing the report somehow.