iam working with sdn controllers and more spesific with ONOS. When i run a topology iam trying with the help of iperf tool to take some measurements between two nodes at TCP mode, as far as bandwidth is concerned, and declare it to have a maximum value for example 2Gbits/sec.
The measurements between nodes have a duration of one minute every 5 seconds. I used the ''-b'' flag but it keeps the bandwidth constant, while I want 2Gbits/sec to be the maximum. For example the ideal would be to have different values like:
1.98
1.89
1.95
2.00
...
...
...
I am attaching the measurements I am trying to take to see if anyone can help me.
[enter image description here]
(https://i.stack.imgur.com/6g640.png)
Not really. You can try the --near-congestion option or -b 1.8g,200m to set vbr. It's in the man page.
https://iperf2.sourceforge.io/iperf-manpage.html
Or use the flows code and step it that way. Flows requires passwordless ssh and python3's asyncio.
I am in the middle of a project where two ADS1220 are needed to configure in order to measure three bridge circuits with differential modes. I am using the MSP430f5529 launch pad to configure ADS1220. I tried to set up it with Arduino and was partially successful as some random numbers were popping up. Does anyone have experience dealing with ADS1220?
Thank you in advance
I am using the latest Grafana 8.2.1 with Amazon Timestream 3.1.1 datasource plugin.
I have noticed that when I use more than one query, the graph will jump / flicker on refresh.
I have the details reported in issue 40424
Just wondering if anyone else has experienced the same thing and if there is a workaround for this?
Panel is trying to optimize axe min/max values based on graphed values. Panel is getting those 2 timeseries in different times (executionFinishTime difference is 6 ms), so that is causing flickering.
I would set static Y-min and Y-max values (e.g. 20, 140 for showed values) to minize that auto optimization of axes. Or you can play also with soft Y-min/max.
Yes, I have flickering problems too. Very bothersome. In my case I perform various readings alternatively every fraction of a second to avoid serial port congestion. Maybe grafana doesn't like the slight difference in time between different measurement series.
I have graphite+collectd setup to collect system related metrics. This question concerns with the memory plugin for collectd.
My infra has this format for collecting memory usage data using collectd:
<cluster>.<host>.memory.memory-{buffered,cached,free,used}
I want to plot the percentage of memory used for each host.
So basically, I have to do something like this:
divideSeries(sumSeriesWithWildCards(*.*.memory.memory-{buffered,cached,free},1),sumSeriesWithWildCards(*.*.memory.memory-{buffered,cached,free,used}),1)
But I am not able to do this, as divideSeries wants the divisor metric to return only one metric.
I basically want a single target to monitor all hosts in a cluster.
How can I do this?
try this one:
asPercent(host.memory.memory-used, sumSeries(host.memory.memory-{used,free,cached,buffered}))
you'll get a graph of memory usage in percents for one hosts. Unfortunately I wasn't able to make it work with wildcards (multiple hosts).
Try this for multiple nodes with regex.
alias(asPercent(sumSeries(collectd.nodexx*_internal_cloudapp_net.memory.memory.used), sumSeries(collectd.nodexx*_internal_cloudapp_net.memory.memory.{used,free,cached,buffered})),"Memory Used")
alias(asPercent(sumSeries(collectd.nodexx*_internal_cloudapp_net.memory.memory.{cached,buffered}), sumSeries(collectd.nodexx*_internal_cloudapp_net.memory.memory.{used,free,cached,buffered})),"Memory Cached")
alias(asPercent(sumSeries(collectd.nodexx*_internal_cloudapp_net.memory.memory.free), sumSeries(collectd.nodexx*_internal_cloudapp_net.memory.memory.{used,free,cached,buffered})),"Memory Free")
I am currently trying to find the method or mechanism for calculating the distance between two wireless mobile nodes.
Let's say, there are 2 nodes (node B and C) coming in communication with node A. How can node A knows whether node B or node C is nearer than the other?
Is it possible to find it out by using RSSI? or are there any other mechanisms to calculate the distance?
I would appreciate any of your answers or comments.
Thank you in advance.
How about using GPS? Most mobile devices have internal GPS support these days.
The RSSI will give you some indication of distance, but signal may have bounced around and penetrated several walls, so a stronger signal doesn't necessarily mean they are closer.
Another option is to use WiFi geolocation, which works pretty good assuming there are several other WiFi access points in the area. I use SkyHook and it works great:
http://www.skyhookwireless.com/location-technology/sdk.php