Hi
I am ingesting a Kafka topic to Pinot and able to run queries on the
datasets using PQL.
However, I am skeptical about the segments created by Apache Pinot which is
of course due to my lack of understanding on Pinot time series segments.
The number of segments created by Pinot is just 3 whereas Druid creates
over 900 segments over the same datasets.
For time series segments, I have chosen a column of type
string("14-AUG-2021 18:05:16") from upstream kafka topic in the
`dateTimeFieldSpecs` section of schema definition:
```
"dateTimeFieldSpecs": [
{
"name": "LAST_UPDATE_DATE",
"dataType": "STRING",
"format": "1:HOURS:SIMPLE_DATE_FORMAT:dd-MMM-yyyy HH:mm:ss",
"granularity": "1:HOURS"
}
]
```
The segment configuration section is:
```
"tableType": "REALTIME",
"segmentsConfig": {
"timeColumnName": "LAST_UPDATE_DATE",
"timeType": "HOURS",
"schemaName": "userRole",
"replicasPerPartition": "1"
},
...
...
```
When I import the resultant table in Apache Superset, I don't find any time
column. This leads me to assume that I am missing something on
segmentsConfig or dateTimeFieldSpecs config specification.
How can I make sure that Pinot creates segments based on specific datetime
columns ?
Thanks
