Recommended Posts

  • LogicMonitor Staff

Time after (Up)time

 

It’s one of the days that an unsung hero gets his chance to make a mark on the world.And who might this be?

The floor function [x], also called the greatest integer function or integer value (Spanier and Oldham, 1987), gives the largest integer less than or equal to x.

 

The floor function is not-normally implemented in the LM perspective – often brushed off, but it got its day when we had a user on chat who was looking on creating a dashboard which has server uptime in days instead of seconds. And so I tried devising a Big Number Widget that would include a virtual datapoint in it with the following calculation to display seconds into days:

 

 

Support Case Article.png

Fig1. Initial UptimeDays setup without the floor() function.

 

 

UptimeSeconds/60/60/24

 

Where UptimeSeconds references a pre-calculated Complex Datapoint from the WinSystemUptime Datasource.

 

 

 

Alas – this was not presenting days in a helpful manner. It was displaying days with a decimal point.

 

Quoting from the chat : that uptime display widget you created would be grand if it showed the days. it says 0,4 at the minute, does that meant 0.4 days?”

 

And so we needed to figure on another method.

A colleague  of ours,  a wizard from the magical realm of complex datapoints pointed out to us something powerful, that could display seconds, minutes and days even –  And, with this, the second hand unwinds...

 

By using the floor() function appended to the original expression – we could actually calculate:

The day rounded to the nearest lowest integer.

 

 

 

floor(UptimeSeconds/ 60/ 60 /24)


 

For example if we had the result of 2.4 days from the UptimeSeconds/6000/60/24 calculation – it will present us the result as 2 – to the smallest following integer.

 

Support Case Article Uptime Datasource.png

 

Fig2. Complex Datapoints for Days, Hours and Minutes with the floor function

 

 

And on the Hours Datapoint:

 

 


 

floor((UptimeSeconds-(days*86400))/3600)


 

Where 86400 represents the number of seconds in a day – 60 * 60 *24 = 86400.

and 3600 represents the number of seconds in an hour - 60 * 60 =3600.

 

 

 

This calculates the number of total hours (excluding the amount already converted and apportioned into the Days metric) , and again the floor function rounds the number hours to the smallest following integer.

For example - 2.4 days - 0.4 is then carried over to the hours, which equates to 9.6 hours, and with the application of the floor function it will reflect 9 hours, with the balance of 0.6 hours to be calculated on the Minutes Datapoint.

 

 

 

And lastly on the Minutes Datapoint:


 

floor((UptimeSeconds-(days*86400)-(hours*3600))/60)


 

Where 86400 represents the number of seconds in a day – 60 *60 *24 = 86400.

and 3600 represents the number of seconds in an hour - 60 * 60 =3600.

and 60 represents the number of seconds in a minute.

 

 

This does the same again, number of total minutes excluding those previously factored in the days (*24) and and the hours (*60), rounded to the smallest following integer.

And therefore, the floor() function has been very useful in this case to catch the Days, Hours and Minutes, have them precisely presented on the Big Number Widget - which the team could display on the dashboard to wait on -- Time after Time.

 

Support Case Article Big Number Widget.png

 

Fig 3. Uptime Days,Hours and Minutes on the Big Number Widget

 

 

References:

 

 

https://www.logicmonitor.com/support/datasources/creating-managing-datasources/datapoint-expressions/

 

http://mathworld.wolfram.com/FloorFunction.html

 

Credits to David Lee for pointing out on the powerful capabilities of the floor function.

 

 

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.