Different Graphs on Reports show different results

Votes:

0

Your Vote:

Up

Down

Hi,

When I run a monthly report on my packetsniffing sensor I get two graphs, one of the month and one for two months. The two graphs show different results. The one for the past month shows much greater traffic then the two month graph. I'm using the pre-built monthly graph.

Any idea why this is?

graphs prtg reports

Created on Dec 28, 2010 2:09:57 PM by  chlhardy (0) 1



4 Replies

Votes:

0

Your Vote:

Up

Down

Hello,

this is most likely due to the different averages used in the two graphs.

Best Regards.

Created on Dec 28, 2010 2:13:38 PM by  Torsten Lindner [Paessler Support] (19,480) 3 1



Votes:

0

Your Vote:

Up

Down

Hi, I'm not sure if I'm following you. It almost appears like two completely different graphs.

Are the calculations used on the graphs different?

Created on Dec 28, 2010 2:20:29 PM by  chlhardy (0) 1



Votes:

0

Your Vote:

Up

Down

As said, you most likely set a different report-averages (interval) for the 1 month graph, then is configured for the 2 month-graph below, and therefore the calculations are different.

Created on Dec 28, 2010 2:55:35 PM by  Torsten Lindner [Paessler Support] (19,480) 3 1



Votes:

0

Your Vote:

Up

Down

The averaging interval is the key. Your sensors each have a configurable monitoring interval and will store the results of the sensor every time the interval has elapsed (30s, 60s, 5m, etc.). When you graph results, if you select an average interval higher than the monitoring interval configured what happens is you take all of the values in the graph interval and average them out. See the following example:

CPU sensor polls every 30s: 2:00:00 - 25%, 2:00:30 - 27%, 2:01:00 - 33%, 2:01:30 - 26%, 2:02:00 - 45%, 2:02:30 - 47%, 2:03:00 - 25%, 2:03:30 - 27%, 2:04:00 - 85%, 2:04:30 - 25%, 2:05:00 - 26%, 2:05:30 - 26%

Looking at these results, you can see there are a couple spikes to 45%-57% and 85%. Given a 5 minute averaging interval, however, you end up with 36.5% for this period which is not representative of the real data that you see above. If you instead averaged by 1min/60sec you're averaging every two values out which gives you a clearer picture but also a LOT more data is processed.

The way I see it, larger intervals are great for viewing utilization over time or for reviewing usage over a very long period. If you're looking for data spikes, though, lower the averaging interval and be more selective about your timeframe or you'll end up twiddling your thumbs for 25 min while you wait for that graph to come back.

Created on Dec 29, 2010 11:54:46 PM by  Blake Webb (0) 1



Please log in or register to enter your reply.


Disclaimer: The information in the Paessler Knowledge Base comes without warranty of any kind. Use at your own risk. Before applying any instructions please exercise proper system administrator housekeeping. You must make sure that a proper backup of all your data is available.

PRTG
Network Monitor
Intuitive to Use.
Easy to manage.

150.000 administrators have chosen PRTG to monitor their network. Find out how you can reduce cost, increase QoS and ease planning, as well.

Visit
www.paessler.com

What is this?

This knowledgebase contains questions and answers about PRTG Network Monitor and network monitoring in general. You are invited to get involved by asking and answering questions!

Learn more

Top Tags


View all Tags