PDE Question 72

Community Discussion

Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.

• 1

  On so many of these questions, how do you actually know if you're correct. I said C but the correct answer was A. Honestly, it's driving me crazy.

• Chose C1

  For the ones saying BigTable is cheaper, BigTable in eu-north1 costs $0.748/hour per node. So if you were to run the node 24/7 you would pay more than 500$ per month. Querying 1TB of data in BigQuery is 7.5$. With smart querying and good database design you can minimize the bytes processed by BQ. So even though BigTable does not directly charge for querying, it charges for running the cluster and the overall price does not make sense. And as far as I know, it's not possible to spin up and shut down BigTable automatically. Also, since the table is an external table to BigQuery, we incur no cost for storing that data in BigQuery and paying 300$ per month for storage.

• 2

  C. Use Cloud Storage for storage. Link as permanent tables in BigQuery for query. Cloud Storage is a highly durable and cost-effective object storage service that can be used to store large amounts of text files. By storing the input data in CSV format in Cloud Storage, you can minimize costs while still being able to query the data using BigQuery. BigQuery is a fully-managed, highly-scalable data warehouse that allows you to perform fast SQL-like queries on large datasets. By linking the Cloud Storage data as permanent tables in BigQuery, you can enable multiple users to query the data using multiple engines without the need for additional compute resources. This approach would be the most cost-effective for querying aggregate values for multiple users, as BigQuery charges based on the amount of data scanned per query, so the more data you store in BigQuery the less you pay per query.