[THIN] Re: How can I test data transfers to my citrix servers ?
- From: "Joe Shonk" <joe@xxxxxxxxxxxxxxxxxxx>
- To: <thin@xxxxxxxxxxxxx>
- Date: Sun, 5 Dec 2004 10:21:04 -0700
Setup your environment to test the same as it will be in production...
Cross connecting servers will give you bandwidth with little/no latency, but
it is unrealistic to benchmark (unless your Intel trying to show your card
is better than broadcom) since you'll servers will be connected to a switch
of some sort.
Same goes with Raid 0 and 0+1. Raid 0 will give you throughput than RAID 5
at the expense fault tolerance. As far as 0+1, you're better off with 1+0.
UNC path vs map a drive. Why would there be a difference? They are both
going to same location and using the same redirector. A mapped drive letter
is essentially a mount point for a UNC path. Yes, drive letters are legacy,
but they can be easily changed (via login script) to point to a new location
where as hard coded UNC paths are a little more difficult to change. Also,
there are still quite a few applications that don't understand UNC paths.
If you want to find what your bottleneck is, I would suggest using perfmon.
Joe
-----Original Message-----
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf
Of Andrew Cooke
Sent: Sunday, December 05, 2004 5:25 AM
To: thin@xxxxxxxxxxxxx
Subject: [THIN] How can I test data transfers to my citrix servers ?
I'm sure at one time or another many of us have wanted to TEST and answer
these questions.
How long does it take (accurately in both time and speed / second)to copy a
50Mb file from the file server to the citrix server ?
Is is faster to map a drive or use a UNC path ?
Once the transfer is finished identify where the limitation was. ie.
bandwidth, NIC speed, raid level, number of users on system at a time, core
switch etc. And of course was it the CLIENT retriving or the SERVER that was
the limitation.
I had the following ideas, but there must be an article where someone
professional has done this before.
1. Connect a cross over cable between the servers and test speed (ie no
users and best possible link speed 1Gb full duplex
2. Change the raid level from RAID5 to RAID0 to RAID0+1 do the same and see
how much difference it makes
Looking for tool that will help me, I want to avoid testing with FTP because
thats not what users are using on a daily basis !
********************************************************
This Weeks Sponsor Activaeon.com
Reduce licensing costs with activAeon XA and
get one month completely free.
http://www.activaeon.com
**********************************************************
Useful Thin Client Computing Links are available at:
http://thin.net/links.cfm
ThinWiki community
http://www.thinwiki.com
***********************************************************
For Archives, to Unsubscribe, Subscribe or
set Digest or Vacation mode use the below link:
http://thin.net/citrixlist.cfm
********************************************************
This Weeks Sponsor Activaeon.com
Reduce licensing costs with activAeon XA and
get one month completely free.
http://www.activaeon.com
**********************************************************
Useful Thin Client Computing Links are available at:
http://thin.net/links.cfm
ThinWiki community
http://www.thinwiki.com
***********************************************************
For Archives, to Unsubscribe, Subscribe or
set Digest or Vacation mode use the below link:
http://thin.net/citrixlist.cfm
Other related posts:
