Imaging much slower than expected.

[laserblazer]

To date been using Ghost 11 for imaging and if I did a peer to peer image I'd get a transfer rate of around 300MB/min. We've just upgraded all of the switches to 1GB in preparation for a server and clients upgrade and I just tried a peer to peer image, expecting it to be 10 times as fast and it's 10 times as slow! I've tried pinging with the buffer at 65500 and that's coming back at <1ms, which seems to indicate that the network is faster, so what's happening? The server is still running at 100mb.

 

I'm hoping that this is a quirk of Ghost and that when we deploy using WDS, it will be fine otherwise it's going to be a long job.

Edited March 29, 2013 by laserblazer

[6Foot2]

We found that Ghost was a real pain in the end [Version: Ghost 8] We had to isolate rooms from the rest of the network to image even one station. WDS does not have this problem [for us at least] On a reasonable spec machine Windows is deployed in less than 10 minutes [applications deployed via GPO/MSI later]

 

Server backbone: 1 Gb. Client connection: 100 Mb.

[sted]

have you got spanning tree or igmp set on switches that can kill ghost

[laserblazer]

have you got spanning tree or igmp set on switches that can kill ghost

 

The LEA fitted the switches yesterday and I haven't had a chance to see how they are configured but I wouldn't have thought so. Oh well, the LEA experts can sort out any problems.

[staningrimsby]

That could be your problem, if your LEA is anything like ours they wont be right.

[cpjitservices]

Have you got multicasting enabled on your switches ? I've found that having this disabled slows down imaging, I have seen it on ghost - WDS doesnt seem to have the issue. I think I've seen this issue on FOG too.

[Duke5A]

Are you using the DOS version of Ghost or Ghost32? The DOS version was always considerably slower and the Win32 counterpart because of the drivers it was sitting on top of. You can boot BartPE or WinPE and launch Ghost32 through there and watch it fly.