Form
We will use all the knowledge learned in this week. Please answer your questions via form above.
Deadline for the homework is 7th Feb 2022 17:00 CET.
What is count for fhv vehicles data for year 2019
Can load the data for cloud storage and run a count(*)
How many distinct dispatching_base_num we have in fhv for 2019
Can run a distinct query on the table from question 1
Best strategy to optimise if query always filter by dropoff_datetime and order by dispatching_base_num
Review partitioning and clustering video.
We need to think what will be the most optimal strategy to improve query
performance and reduce cost.
What is the count, estimated and actual data processed for query which counts trip between 2019/01/01 and 2019/03/31 for dispatching_base_num B00987, B02060, B02279
Create a table with optimized clustering and partitioning, and run a
count(*). Estimated data processed can be found in top right corner and
actual data processed can be found after the query is executed.
What will be the best partitioning or clustering strategy when filtering on dispatching_base_num and SR_Flag
Review partitioning and clustering video.
Clustering cannot be created on all data types.
What improvements can be seen by partitioning and clustering for data size less than 1 GB
Partitioning and clustering also creates extra metadata.
Before query execution this metadata needs to be processed.
In which format does BigQuery save data
Review big query internals video.