Is there any offline analytical processing
? If there is, what is the difference with online analytical processing
?
In What are OLTP and OLAP. What is the difference between them?, OLAP is deals with Historical Data or Archival Data. OLAP is characterized by relatively low volume of transactions. Queries are often very complex and involve aggregations.
I don't understand what does the word online
in online analytical processing
mean?
Is it related to real time processing
(About real-time
understanding: In a short time after the data is generated, this data can be analyzed.
Am I wrong with this understanding)?
When does the analysis happen?
I imagine a design like this:
log generated in many apps -> kafka -> (relational DB) -> flink as ETL -> HBase, and the analysis will happen after data is inserted into HBase. Is this correct?
If yes, why is it called online
?
If no, when does analysis happen? Please correct me if this design is usually not in the industry.
P.S. Assuming that the log generated by the apps in a day has a PB level