| View previous topic :: View next topic |
| Author |
Message |
swingkyd
Guru
Joined: 13 Jan 2005
Posts: 334
|
 Posted: Mon Oct 29, 2007 9:12 pm Post subject: question about portage and rsync from home |
 |
|
This isn't really a "problem," rather just a question for people: I am running an rsync service for my local computers to sync their portage to one of my machines. I noticed that the output after syncing to my local computer is something like this:
| Code: | Number of files: 128996
Number of files transferred: 2188
Total file size: 171343235 bytes
Total transferred file size: 7950555 bytes
Literal data: 7950555 bytes
Matched data: 0 bytes
File list size: 3129952
File list generation time: 1.269 seconds
File list transfer time: 0.000 seconds
Total bytes sent: 54703
Total bytes received: 6049123
sent 54703 bytes received 6049123 bytes 20177.94 bytes/sec
total size is 171343235 speedup is 28.07
|
What I was wondering about was the number 20177.94 bytes/sec. Is it true that I'm only transfering at <20k/sec? I would imagine that syncing on a 100Mbit lan (all wired) would be much faster? Or is it being slowed down by my computer somehow? Or am I totally out to lunch?
thanks for your input. |
|
| Back to top |
|
 |
Genone
Retired Dev
Joined: 14 Mar 2003
Posts: 9532
Location: beyond the rim
|
| Posted: Mon Oct 29, 2007 9:24 pm Post subject: |
|
|
| You should be able to answer that yourself: The sync transfered about 6MB of data, so at 20 kb/s it would have taken about 5 minutes. And if that's the case note that rsync probably doesn't exclude phases where it didn't transfer anything from the calculation (e.g. generating the filelist, or comparing mtimes or checksums), so a slow CPU or disk could have a significant effect on the time (and therefore the transfer rate). |
|
| Back to top |
|
|
swingkyd
Guru
Joined: 13 Jan 2005
Posts: 334
|
| Posted: Mon Oct 29, 2007 9:49 pm Post subject: |
|
|
thanks... I wasn't sure if it computed it based on actual transfer time or like you say, total time to perform all operations. It doesn't really make much sense to report total time to perform and then report bytes per second when it's a totally skewed result.
Basically, I am more concerned about any potential problems with my network than with computation or disk IO time.
But thanks for your response. |
|
| Back to top |
|
|
|
Portage & Programming
