topic Re: FabricPool/Inactive Data Reporting in ONTAP Discussions

FabricPool/Inactive Data Reporting

TAO1 — Thu, 22 Sep 2022 08:45:10 GMT

Hi,

I'm looking into FabricPool for a supported FAS platform. The intention is to use the maximum cooling threshold of 183 days. Before I go down the route of enabling FabricPool, I want to run an analysis on the data to establish how much data qualifies as cold in relation to the 183 day cooling period. The documentation says:

For an aggregate that is not used for FabricPool, inactive data reporting uses the tiering policy to decide how much data to report as cold.
- For the none tiering policy, 31 days is used.
- For the snapshot-only and auto, inactive data reporting uses tiering-minimum-cooling-days.
For the ALL policy, inactive data reporting assumes the data will tier within a day.

Until the period is reached, the output shows “-” for the amount of inactive data instead of a value.

Am I really going to have to wait 183 days before I can establish how much data qualifies as cold?

Thanks

Re: FabricPool/Inactive Data Reporting

TAO1 — Fri, 23 Sep 2022 10:20:20 GMT

Anyone from NetApp?

Re: FabricPool/Inactive Data Reporting

TAO1 — Mon, 26 Sep 2022 08:15:22 GMT

Anyone?

Re: FabricPool/Inactive Data Reporting

TAO1 — Wed, 28 Sep 2022 10:02:51 GMT

Is this a dumb question?

Re: FabricPool/Inactive Data Reporting

Ontapforrum — Wed, 28 Sep 2022 11:04:47 GMT

It's not a dumb question at all (who cares). I find it interesting and tough to answer.

I am just trying to give some perspective to your query. Hopefully, experts will pitch in. You are anyway more than welcome to raise a ticket and put this question through. You can even put this question to your NetApp Account manager, and he will manage the reply for you.

My comments: Since data tiering is done at the "block level", a data block is considered cold after it hasn’t been accessed for a certain period of time, which is defined by the tiering policy’s minimum-cooling-days attribute. It doesn't care about the file, so I am guessing it is not going to tell you how old the file is based on the file's 'access/modify' time. Also, there is no visibility as which data type is cold. It is simply going to tell you the percentage of blocks that are cold. I believe once the scan is completed, it is just going to tell you how much '%' or 'mb/gb/tb' is cold, now whether it will report this after the configured cooling period has passed (irrespective of how old the files are) is something I am also unsure, but you can reach out to the channels I suggested.

However, if I read these lines from the following pdf, it clearly seconds what you have already stated.

https://docs.netapp.com/us-en/cloud-manager-tiering/faq-cloud-tiering.html

"IDR starts showing information after the configured cooling period has passed. Using ONTAP 9.7 and earlier, IDR had a non-adjustable cooling period of 31 days. Starting with ONTAP 9.8, the IDR cooling-period can be configured up to 183 days".

Kb for reference:
https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/How_does_the_FabricPool_inactive_data_reporting_work

Re: FabricPool/Inactive Data Reporting

TAO1 — Tue, 04 Oct 2022 15:04:16 GMT

Thanks.

Of course, tiering is based on blocks as opposed to reading file property attributes. I'm assuming that when IDR is enabled, only then does it start registering the frequency that blocks are read