Skip to Content.
Sympa Menu

wg-multicast - RE: Lab imaging using multicast

Subject: All things related to multicast

List archive

RE: Lab imaging using multicast


Chronological Thread 
  • From: "Quigley, Mark R" <>
  • To: "Ying Zhang" <>
  • Cc: "wg-multicast" <>
  • Subject: RE: Lab imaging using multicast
  • Date: Tue, 6 May 2008 12:11:18 -0700

In a past life, we had a similar problem at University of Idaho with
Novell servers imaging clients. Unicast 20 min; multicast 1.5 hrs plus.

Background: Novell servers with 1Gb nics connected to Cisco 6500, 100Mb
client switches with 100Mb uplink from older 6126-100FX blades (64KB
port buffer).

Testing:
- Multicast from 1Gb server to 100Mb uplink switch clients trouble.
- Multicast from 100Mb server to 100Mb uplink switch clients okay.
- Multicast from 1Gb server to 100Mb uplink switch clients (using
10/100/1000 blade with larger port buffers ~= 5MB) okay.

Diagnosis: Multicast from 1Gb servers resulted in numerous discards when
client switch uplink was on 6124-100FX ports. When buffer size was
adequate, discards did not happen. Multicast imaging time around 30 min.

Recommendation: update client switches to 1gb uplink.

Hope this helps,
Mark

-----Original Message-----
From: Ying Zhang
[mailto:]

Sent: Tuesday, May 06, 2008 7:14 AM
To:

Cc: wg-multicast
Subject: Re: Lab imaging using multicast

We use Nortel ERS5500 switches.

Ying
----- Original Message -----
From:
<>
To: "Ying Zhang"
<>
Cc: "Yongrui Tang"
<>;
"Dan Oachs"
<>;
"wg-multicast"
<>
Sent: Tuesday, May 06, 2008 11:09 AM
Subject: Re: Lab imaging using multicast


> Hi,
>> Thanks for the reply. We'll double check that. When we set up a test
>> environment with three clients, they all run at 1-gig full-duplex. It

>> took 30 minutes to image all three with unicast but 75 minutes with
>> multicast.
>
> what switches? some makes have issues with eg IGMP awareness being
> turned on - performance can go through the floor(!)
>
> alan
>




Archive powered by MHonArc 2.6.16.

Top of Page