Dr. Elaina's GCP Challenge: Do you know your GCP?

Edit Answers
123456789101112131415161718192021222324252627282930313233
Across
  1. 2. Delivery method that is ideal for subscribers needing close to real time performance
  2. 4. Pub/sub connects applications and services through a ______ infrastructure
  3. 7. Dataflow models find out when in processing time results are materialized via ____, triggers, and allowed lateness
  4. 12. In a streaming system on GCP this allows you to query data as it arrives from the streaming pipelines
  5. 13. In Dataproc, they cost less but may not always be available
  6. 14. A machine learning problem where the outcome to be predicted is a continuous number
  7. 17. Is a dataflow set of data in your pipeline
  8. 19. Is a class that will do logistic regression
  9. 23. BigQuery supports nested and ______ fields
  10. 24. Dataflow models find out how refinements of results relate via _______ modes
  11. 26. ML API can ___ scanned receipts
  12. 29. Determines the Google data centre where compute nodes will be
  13. 30. A _____ combines its inputs to map part of a decision surface in a neural network
  14. 31. Bigtable learns access patterns and attempts to distribute reads and storage across _______ evenly
  15. 32. Is a dataflow processing operation or step in your pipeline
  16. 33. Provides the ability for spark programs to seperate compute and storage
Down
  1. 1. Performance of this can be improved by changing schemas to minimize data skew
  2. 3. Is a dataflow endpoint for your pipeline
  3. 5. Installing software libraries on the master in dataproc typically uses an ______ script
  4. 6. Dataflow models find what results are calculated via ________
  5. 8. Enhancing ML data by extracting features from raw data is feature______
  6. 9. In a neural network the number of layers is a _______
  7. 10. One complete pass through the training dataset in Machine Learning
  8. 11. An efficient way to read data into TensorFlow
  9. 15. Dataflow models find where results are calculated via event-time _______
  10. 16. A small set of examples on which gradient is computed in Machine Learning
  11. 18. Bigtable is great for __________ data
  12. 20. Lets you train TensorFlow machine learning models at scale
  13. 21. This cluster mode in Dataproc provides 1 master and N workers
  14. 22. The best practice for querying a table, then querying the results of that query is to use a _____
  15. 25. A software framework for writing portable ML code
  16. 27. Makes it easy to create resilient streaming pipelines
  17. 28. Dataproc helps you create job-specific clusters without ______