I work in an office with about 30 employees, and we have a bonded T1 pair that's used for both our VOIP phone system and our internet access. Our VOIP system takes top priority, which leaves us with just a little bit under 3 Megabits/second for internet bandwidth. (And yes, when multiple employees try to use the internet at once, it's every bit as awful as you might imagine.)
All our internet traffic goes through an ASG-220 currently running UTM 9.104-17. Back when we had UTM 8 installed, the traffic usage graphs, either in the Executive Reports or in the web interface, were very useful. They showed numerous long plateaus throughout the day, where the "flat spot" at the top was just below 3 Mb/s. The plateaus were even more obvious in the weekly, monthly, and yearly graphs. It was quite clear from the graphs that we were frequently using every bit of bandwidth we had available, and I was using these graphs to help justify to my management that we needed to upgrade our internet access.
Since upgrading to UTM 9, however, the plateaus are gone. The network traffic graphs now show only sharp peaks, that only rarely rise to 3 Mb/s. Here's an example:
Attachment 10544
Last week, one of our T1 lines went down, so we had only 1.5 Mb/s for a couple of days, and the difference was barely even visible on the graphs, though it was obvious from the complaints I got from users about "slow internet". When my management looks at the UTM 9 graphs, their response is that we're clearly not using all our bandwidth, so why should they pay for more?
Did some graphing algorithm or parameters change in UTM 9, and if so, is there any way to change them back to work like they did in UTM 8? Or do I have some kind of setting wrong somewhere, and if so, what could it be? I know I can saturate a bonded T1 pair all by myself without even trying hard, so why don't I see it in the usage graphs? I should note that the ASG does not have QOS turned on. And when I run network speed tests after hours, I typically do get upload and download speeds in the range of 2.9 Mb/s, so I think we're getting what we're supposed to have.
Can anyone explain?
All our internet traffic goes through an ASG-220 currently running UTM 9.104-17. Back when we had UTM 8 installed, the traffic usage graphs, either in the Executive Reports or in the web interface, were very useful. They showed numerous long plateaus throughout the day, where the "flat spot" at the top was just below 3 Mb/s. The plateaus were even more obvious in the weekly, monthly, and yearly graphs. It was quite clear from the graphs that we were frequently using every bit of bandwidth we had available, and I was using these graphs to help justify to my management that we needed to upgrade our internet access.
Since upgrading to UTM 9, however, the plateaus are gone. The network traffic graphs now show only sharp peaks, that only rarely rise to 3 Mb/s. Here's an example:
Attachment 10544
Last week, one of our T1 lines went down, so we had only 1.5 Mb/s for a couple of days, and the difference was barely even visible on the graphs, though it was obvious from the complaints I got from users about "slow internet". When my management looks at the UTM 9 graphs, their response is that we're clearly not using all our bandwidth, so why should they pay for more?
Did some graphing algorithm or parameters change in UTM 9, and if so, is there any way to change them back to work like they did in UTM 8? Or do I have some kind of setting wrong somewhere, and if so, what could it be? I know I can saturate a bonded T1 pair all by myself without even trying hard, so why don't I see it in the usage graphs? I should note that the ASG does not have QOS turned on. And when I run network speed tests after hours, I typically do get upload and download speeds in the range of 2.9 Mb/s, so I think we're getting what we're supposed to have.
Can anyone explain?