Hi,
I'm trying to calculate bandwidth usage for our Citrix environment, and I'm not sure if what I'm doing is accurate, and if there's an easier, more reliable way. (I'm doing this because I want to move from Secure Gateway to Netscaler VPX, and want to make sure I buy the correct one, or use the VPX Express).
Here's the situation:
Citrix Xenapp, using Web Interface, Secure Gateway.
Sonicwall with secure gateway in dmz (interface 2, let's say).
Everyone connects through secure gateway (internal and external users - I know, I know)
In the sonicwall, I see interface statistics, e.g. rx bytes, tx bytes, and others (unicast packets, etc.)
Since all Citrix traffic goes through the Secure Gateway, I figure I can use the rx bytes and tx bytes on the dmz interface, reset statistics, and take samples every 5 or 10 minutes. (The stats are cumulative and not per any time period.)
I then do calculations in a spreadsheet, as follows:
sample period - 5 minutes
rx bits = rx bytes * 8
rx bits per minute = rx bits / 5
rx bits per second = rx bits per minute / 60
rx mbps = rx bits per second * 1024 * 1024
Same calculations for tx.
Add the rx mbps and the tx mbps, and that's my total mbps for Citrix traffic. (Note, there's not much else running through the DMZ, so I feel confident that 99% of the traffic is Secure Gateway traffic.
When I implement Netscaler, would it be reasonable to use my calculated bandwidth methodology?
Also, is there a better / easier way to do this? If i need to install something on the Secure Gateway, I can do that. If I need to purchase something for the Sonicwall, I might do that. I have an NSA series Sonicwall.