Could someone help me validate if I am collecting the reports correctly (attribution-report NODE JS version) about aggregation-service HOT 4 CLOSED

Hi @dbrito ,

This looks about right. I see that your payload is base64 decoded and converted to a bytestring. And the key_id and shared_info are strings. Converted into the aggregation service report avro schema.

from aggregation-service.

@maybellineboon thanks!
Just one question, the part where each "aggregation_service_payloads" in the report becomes an .avro file is also correct, right?

In this case, there is no scenario where I will receive more than one "aggregation_service_payloads" per report and I have to store these records in the same .avro file, right?

from aggregation-service.

hi @dbrito ,

You can either collect all the json reports and convert them into a single file or you can create separate avro reports. For better performance, it is recommended to split the number of reports to the number of CPUs for your Aggregation Service instance. For example, you have 1,000 reports with 6 CPUs, you will have 6 files and the 1,000 reports will be split into 6 files.

Also, when batching, we do recommend to follow the batching strategies doc.

from aggregation-service.

Thanks !
I'll take a look

from aggregation-service.