Add page for QoSHistogram object

[scaleoutsean]

Page URL

https://docs.netapp.com/us-en/element-software/api/reference_element_api_listvolumeqoshistograms.html

Page title

ListVolumeQoSHistograms

Summary

I think this page has several problems:

a) ListVolumeQoSHistograms "generates a histogram of volume QoS usage for one volume or multiple volumes." But not all objects are related to QoS itself. For example, writeBlockSizes and readBlockSizes merely describe the workload.

b) There's no information about the time interval over which such values are collected.

Given a belowMinIopsPercentages histogram like this:

"belowMinIopsPercentages": {
  "Bucket1To19": 100,
  "Bucket20To39": 100,
  "Bucket40To59": 100,
  "Bucket60To79": 100,
  "Bucket80To100": 0
}

Is this since the volume was created?

The docs say QoS histograms are generated approximately 5 seconds after volume creation, but does not say whether information offered by the API goes back 1 hour or all the way back.

Furthermore, belowMinIopsPercentages values from two volumes will differ, e.g. the one above has (100,100,100,100,0) while another may have (500,500,500,500,0).

In such situation, can one say that these two volumes comparable belowMinIopsPercentages is used or not? What is the difference between (1,1,1,1,0) and (2,2,2,2,0)?

c) Element API documentation for QoSHistogram object does not exist. Rather than answer the above questions I'd prefer that the missing page be created.

We have them for VolumeObject, QoSObject, etc. QoSHistogram is one of the few undocumented ones and I wouldn't ask for it simply because it's missing but as mentioned above ListVolumeQoSHistograms isn't documented nearly well enough to suffice.

Public issues must not contain sensitive information

• [X] This issue contains no sensitive information.

[scaleoutsean]

Perhaps I should also offer this example.

Obviously volume IDs (last column) that are completely idle should have Bucket80To100 (b_80_to_100) completely maxed-out since they spend 100% of their time below Min IOPS.

But that figure is instead 0 (for example, volume ID 37).

name: histogram_below_min_iops_percentages
time                 b_01_to_19 b_20_to_39 b_40_to_59 b_60_to_79 b_80_to_100 cluster volume_id
----                 ---------- ---------- ---------- ---------- ----------- ------- ---------
2024-05-26T16:48:00Z 28         3          2          0          0           PROD    1
2024-05-26T16:48:01Z 32         2          1          0          0           PROD    2
2024-05-26T16:48:02Z 1          0          0          0          0           PROD    37
2024-05-26T16:48:03Z 1          0          0          0          0           PROD    39
2024-05-26T16:48:04Z 0          0          1          0          0           PROD    40
2024-05-26T16:48:05Z 0          0          1          0          0           PROD    42
2024-05-26T16:48:06Z 0          0          1          0          0           PROD    43
2024-05-26T16:48:07Z 1          0          0          0          0           PROD    45
2024-05-26T16:48:08Z 0          0          0          0          0           PROD    108
2024-05-26T16:48:09Z 0          0          0          0          0           PROD    111
2024-05-26T16:48:10Z 0          0          0          0          0           PROD    113
2024-05-26T16:48:11Z 8          2          7          1          1           PROD    115
2024-05-26T16:48:12Z 18         1          7          3          2           PROD    117
2024-05-26T16:48:13Z 20         0          6          7          2           PROD    120
2024-05-26T16:48:14Z 16         0          9          2          1           PROD    122
2024-05-26T16:48:15Z 17         1          7          3          0           PROD    124
2024-05-26T16:48:16Z 49         1          7          3          1           PROD    125
2024-05-26T16:48:17Z 18095      2257       186        488        1460        PROD    133
2024-05-26T16:48:18Z 875        65         13         16         1           PROD    134
2024-05-26T16:48:19Z 769        50         7          20         6           PROD    136
2024-05-26T16:48:20Z 0          0          2          0          101         PROD    139

[netapp-pcarriga]

@scaleoutsean Hi Sean, thank you for bringing this to our attention. I will investigate and update you here.