Qlik Replicate ADLS Gen2 File Format JSON - Loads taking time

[priyankaraj]

Hello Team,

We are noticing considerable Replicate performance degradation in Replication loads when switching from CSV file format to JSON.

Curious to know why and if someone else has same experience. Thank you!

[john_wang]

What's the Replicate version are you using now. 

I got report that the CPU  utilisation went significantly high when switching from CSV file format to JSON. We are working on it. Will keep this post updated.

[priyankaraj]

Hi John,

Thanks for your quick reply. The CPU utilization was variable 30-40%. I had multiple replication tasks running. 

We are using version 6.6.0.298.  Thank you.

[JitenderR]

@john_wang  Is this targeted to fix in a specific release?

Regards

JR

[priyankaraj]

Hello Team,

Any update? @john_wang 

[john_wang · Accepted solution]

Hi @priyankaraj , @JitenderR ,

Sorry for the delay.

I have not further information about it. I'd like suggest you open a support case in Salesforce and provide necessary information/files for SME/R&D analysis.

Regards,

John.

[priyankaraj]

Hi John,

Any update on this?

[john_wang]

Hello @priyankaraj ,

In the support case (#02042291) the behavior cannot reproduced any more and we cannot get further information about it. I'd like suggest you open a support case with below information if the behavior persists in your env:

1) It occurs during FL or CDC?

2) On which phase exactly the CPU raises - when the file was generated by Replicate?

3) What is the CPU for the same scenario comparing to CSV?

4) Please share the relevant details for reproduction (i.e. table DDL, what is the table size, diagnostic package, how many rows there are in the table).

Regards,

John.