Hello everyone! I would like to introduce my newest plugin to the Grafana community; Psychart: A Psychrometric Chart for Monitoring Data Center Health. A psychrometric chart displays several thermodynamic properties of air-vapor mixtures like temperature and humidity. Primarily these are used in HVAC applications but are sometimes used in meteorological applications or chemical process monitoring. I am introducing this to Grafana because of my team’s need for monitoring physical hardware as well as software. Most of our compute equipment is air-cooled and the air needs to be within a specific range, which can easily be visualized by using a psychrometric chart.
Chart Configuration:
The configuration interface is designed to be easy to use and in a logical order. First, select whether data is measured in US or SI units and then input the elevation of the data center in feet (US) or meters (SI). Leave the dry bulb range about 20-120F or 0-50C and maximum dew point as 90F or 40C to start. The ASHRAE comfort regions are shown in the screenshot below.
Data Configuration:
2 data series must be measured in order to “fix the state” and plot data. That means that the other thermodynamic properties can be calculated after only 2 properties are known. Choose the option of the 2 series measured and use the field drop downs to select them.
Display Configuration:
Psychart also allows you to change the point size, optionally connect points with a line, change the color gradient, and optionally show more advanced state variables when hovering the mouse over the points.
Sample screenshot of Psychart rendering inlet air properties for a compute cabinet. The darkened regions are recommended envelopes of air for various types of data centers and IT spaces which are defined by ASHRAE.
Let me know if you decide to use it, and if you have any questions!
Hi @anek! Thank you for your question. The 2 series of thermodynamic data can come from different sensor hardwares. However, it makes the most sense for the sensors to be physically located near one another. Once this plugin is installed, you can add as many psychrometric charts to your Grafana dashboards as you like. Let me know if this answers your question.
Hi, i’m trying to plot my data (Copy of Google Sheets Datasource - Average Temperature - Google Sheets) and i have configure it. But it only shows ashrae chart and not show any data from my table. Then i see my data with table view, the data exist. Any suggestion how to plot data from each rack into the psychart? Thank you
Hi @ehadaian ! Thank you for installing my plugin and trying it out. Currently, my plugin only supports one data series at a time.
I notice that on your Google Sheet, your timestamps are all the same (July 1, 2023, 8:00 AM) so it only plots one data point. Try to change the timestamps for each data point, and try again, while making sure that your time range in Grafana encompasses those time stamps, like so:
For a multiple-series situation, like yours, I would recommend using multiple panels. Otherwise, you could use 1 panel, with Grafana variables to create a dropdown selection on which rack to view.
Hi, i want ask about this, I use this to count the existing shelves but why isn’t there or is the name of the shelf written here, so it’s difficult for me to search for one point at a time to find the name of the shelf I want to search for, how can we find out the name of the purple dot, please explain thanks.
Hi @ventura , many thanks for your feedback.
So the plot is dependent with its timestamp.
Then can we give attributes on each datapoint so we can see a differentiation aside from timestamp?
@dimasjuliansyah221 Hi and welcome to the Grafana community, and thanks for trying out my plugin! At this time, Psychart is capable of plotting 1 (time-dependent) data series at once.
@ehadaian That is corect; the plots are time-dependent.
What I recommend to both of you at this time, is to create a dashboard variable containing your cabinet or location names, and update your query to use that variable. For example, a variable may be Location and a Prometheus query may look like this:
Wew, this is a great insight. Well i’m newbie here and i can’t find any references after looking around related variables into google sheet datasource.
My data is in google sheets.
Prometheus is another thing. Any suggestion on this one?
As of version 4.0.0 and above, Psychart now has the ability to render multiple data series in the same panel!
Watch the video below [no audio] for a guide on how to set it up.
This will allow you to do things like monitor environmental conditions for each compute cabinet, view entering and exiting conditions from an air handler/conditioner, or any other instance where you would need to view sensor data from several different sensors.
Feel free to reply here if you have any questions.
