I am learning Dagster, so sorry if this is a very basic question. Here is what I am trying to understand in Assets. I have a call to a database or webserver that returns an array of values lets call them customer IDs. For each cutomer ID I want to run an asset to analyze that customer. Here is a pseudo code of the idea:

# Find array of customer_ids to analyze
@asset
def find_customers_to_analyze(context):
    with database.get_resource().get_client() as client:
        detector = CustomerWrapper(context)
        return detector.find_customrs_to_analyze() # returns an array of zero ore more customer_ids


# For each customer_id returned by find_customers_to_analyze
@asset
def analyze_customer(context, customer_id):
    with database.get_resource().get_client() as client:
        detector = CustomerWrapper(context)
        return detector.check_for_anomalies(customer_id)

It would be nice if

analyze_customer

would run as many as possible in parallel as allowed by some concurrent limit. Is there some documentation or tutorial somewhere that describes how to achieve this. I assume this is a very common usecase but my Google-foo is not strong enough to find what I am looking for. I found similar unanswered question os SO

This might be what you are looking for https://docs.dagster.io/concepts/ops-jobs-graphs/dynamic-graphs#dynamic-graphs

Thanks @martin o leary that document uses

ops

Can I do the same thing with assets or must I use ops?

I think of partitions as being “outside the asset” if that makes sense? You could build a dynamic graph as above and create an asset from it maybe? https://docs.dagster.io/concepts/assets/graph-backed-assets#graph-backed-assets