ETL Lead Interview Questions and Answers

Difference between Row_num and rowid in SQL and removing duplicates

• ROW_NUM is a function that assigns a unique number to each row in a result set

• ROWID is a unique identifier for a row in a table

• To remove duplicates, use the DISTINCT keyword in a SELECT statement

• Another way to remove duplicates is to use the GROUP BY clause

• The HAVING clause can be used to filter out duplicates based on a condition

Validation of data between source and target is a crucial step in ETL process.

• Compare the data types and formats of source and target data

• Check for missing or extra data in target

• Verify data transformation rules are applied correctly

• Perform data profiling to identify anomalies

• Use checksums or hash values to ensure data integrity

Teradata architecture is a massively parallel processing (MPP) architecture designed for data warehousing and analytics.

• Teradata architecture follows a shared-nothing approach.

• It consists of parsing engine, BYNET, access module processors (AMPs), and disks.

• Parsing engine receives SQL queries and generates an execution plan.

• BYNET is responsible for interconnecting the parsing engine and AMPs.

• AMPs store and process data in parallel.

• Disks store the actual data in a distributed m...read more

Different types of Lookup in ETL

• Cached Lookup

• Dynamic Lookup

• Unconnected Lookup

• Connected Lookup

• Static Lookup