So my Director has asked me to define several basic resources
In this instance. I am being asked to provide
the total drive space available vs the total drive space used
for Application X that lives on a list of servers.
I get what is being asked. There is a goal to tell the Business users
"This applications uses this much space, this much CPU, memory..etc...etc.."
My first thought was simply to pull the disk size in bytes vs the bytes used
As I look through the API v2,
I did NOT see a metric for "disk size in bytes"
What I did find was
It would follow (I think, please correct me if I am wrong)
That is I had a static point in time of all the "used"
added to the "available" I would have in effect the "total" drive size
So the tricky part is I saw one of 2 ways of doing this
1) Pick a static metric point (say 1 AM.. any day) and pull the stats on all the servers at that point in time only
2) Use the INF setting for resolution
My problem is in the understanding of the term
"One aggregated data point of each series. Set Inf to use this option"
Is this saying that it is taking all the different time series of a metric and adding them then averaging?
Would it be more accurate to simple take option #1 and pick a static point for the day to do the calculation?>
AM I missing anything fundamental?
Thanks for this awesome community!
Solved! Go to Solution.
The metric API v2 typically is used to fetch the complete series of data points over time for a selected metric such as "builtin:host.disk.avail". If you want to get one scalar value representing the complete series as one value (typically as average) you can use the 'Inf' option.
So you would then get one value that represents the average disk available over the selected period of time.