Very slow and long running simple querry

Question

Running this :

from azureml.opendatasets import NycTlcGreen

data = NycTlcGreen()
df = data.to_spark_dataframe()

Display 10 rows

display(df.limit(10))

Azure Synaps : 8 vcpu/64 go (3 nodes ) run for over 30 min / 40 min with no error, no results : had to stop it as the same querry on a box of ( 4 vcpu /16 Go got the results in less than 15 min).

Any idea about what I'm missing here ?

Thank for your help.