Practice Test 3 | Google Cloud Certified Professional Data Engineer | Dumps | Mock Test
Your company uses Google Cloud as its main cloud platform. The data science team is working on building a binary classification model and they choose Apache Spark MLLib to build the model. The training set to be used is over 30GB in size.
As the data engineer, data science team lead asked you to spin up an Apache Spark cluster for data scientists to experiment on. He informed you that the cluster’s local HDFS data is not critical and performance is not an issue until the deployment phase. Which of the following stack will you use?
A. Train data scientists to use Cloud ML Engine for building their classification model.
B. Launch a Dataproc cluster in high-availability mode using high-memory worker machine types.
C. Launch a Dataproc cluster in standard mode using high-CPU worker machine types.
D. Launch a Dataproc cluster in standard mode using high-memory worker machine types.
Correct Answer: D
Option A is incorrect: Cloud ML Engine is used to deploy the model after being built. You cannot implement the model using ML Engine.
Option B is incorrect: The scenario states non-critical experiments will be conducted by data scientists, Dataproc cluster used can be in standard mode.
Option C is incorrect: Same as B, the scenario states non-critical experiments, there is no need for high-CPU worker machine types.
Source(s):
Compute Engine Machine Types:
https://cloud.google.com/compute/docs/machine- types#standard_machine_types
Cloud Dataprep:
Comments are closed, but trackbacks and pingbacks are open.