If you are unable to create a new account, please email support@bspsoftware.com

 

News:

MetaManager - Administrative Tools for IBM Cognos
Pricing starting at $2,100
Download Now    Learn More

Main Menu

Data server connections Load Metadata

Started by sdf, 14 Sep 2018 11:12:16 AM

Previous topic - Next topic

sdf

Hi,

Can anyone shed some light..
Is it required to "Load Metadata"?
Under Data server connections> select connection>properties>schemas


bus_pass_man

No it isn't required, but if you want to see anything which is in a particular schema and make use of it you'd need to load the schema.

sdf

I guess, it is required if we are creating data module from it.
Which leads me to another question.
When we load schema metadata, where is it stored? in the \data directory or content store?
Are there any limit on the number of schema metadata we can load?
Is there a memory issue if say we have hundreds of schema?
I noticed a spike in cpu memory when loading a metadata. I'm sure this is because, the actual loading are sql select statements on the selected tables and views. I am worried since we have lots of schema deployed.

the6campbells

When we load schema metadata, where is it stored? in the \data directory or content store?
ANSWER> In CM

Are there any limit on the number of schema metadata we can load?
ANSWER> No hard limit

Is there a memory issue if say we have hundreds of schema?
ANSWER> Just as with an FM model, don't import clutter that will never be used. Waste of time/resources.

I noticed a spike in cpu memory when loading a metadata. I'm sure this is because, the actual loading are sql select statements on the selected tables and views. I am worried since we have lots of schema deployed.
ANSWER> Set the load options to not "sample" the tables as the schema is introspected.

sdf

Thank you very much for the info. I appreciate it.

QuoteI noticed a spike in cpu memory when loading a metadata. I'm sure this is because, the actual loading are sql select statements on the selected tables and views. I am worried since we have lots of schema deployed.
ANSWER> Set the load options to not "sample" the tables as the schema is introspected.

Can you explain what this Retrieve sample data do? How does this affect the data presented when we run reports/dashboard using a datamodule with this setting?

I can't find any documentation that would explain this setting.

Thanks again!
sdf

the6campbells

Inference of the role/properties of a column are derived from the data type and column name

By issuing a basic query and looking at the values in the first-N rows of the result set, further inference of some "metadata" can be inferred.

Can help things like join advisor, intent based modelling.

But often, you are not loosing much if you simply import schemata and do no "sample" data.

sdf

that made sense..

thanks a lot for the info!

cheers!
sdf