Option 4 is the correct answer. Nearline storage is a low-cost, durable storage service for storing frequently accessed data. It is slightly less available than other storage classes and has a minimum data retention period of 30 days. Nearline storage is the best choice over standard storage if you want to reduce storage costs when storing frequently accessed data.

Option 1 is incorrect. Cold line storage is suitable for objects that are accessed up to once a year. It's more cost-effective than nearline storage, but it's not suitable for frequently accessed data.

Option 2 is incorrect. Archive storage is the lowest cost and most durable storage service for data archiving, online backup and disaster recovery. It's more cost-effective than nearline storage, but it's not suitable for frequently accessed data.

Option 3 is incorrect. Multi-region storage is only available for objects stored in multi-regions or dual regions and is ideal for frequently accessed data (“hot” data) or data that is stored for a short period of time. This is less than optimal as it costs more than nearline storage.

【reference】 https://cloud.google.com/storage/docs/storage-classes

Option 1 is the correct answer. You can get the metadata for the specified project by running the command `gcloud projects describe [project ID]`. Run the command with a specific project in gcloud projects. The describe command is an action that retrieves metadata.

Option 2 is incorrect. The command of gcloud describe projects is incorrect. To be correct, it is necessary to describe the commands in the order of gcloud projects describe.

Option 3 is incorrect. The command for gcloud list projects is incorrect. To be correct, write the commands in the order of gcloud projects list.

Option 4 is incorrect. The command `gcloud projects list` is a command to get a list of projects. You cannot get the metadata for a particular project.

【reference】 https://cloud.google.com/sdk/gcloud/reference/projects/describe

Option 2 is correct because it provides fast response and requires the minimal amount of setup.

Option 1 is not correct because setting triggers in each individual database requires additional setup.

Option 3 is not correct because it requires custom programming.

Option 4 is not correct because it requires significant custom programming.

【reference】

https://cloud.google.com/functions/docs/calling

Nearline Storage is a class of Cloud Storage designed for objects that will be accessed at most once every 30 days. Coldline Storage is suitable for objects accessed at most once per year. Multi-regional storage is best suited for objects that should have low latency access from multiple regions. Persistent disks should not be used for archival storage. For more information, see https://cloud.google.com/storage/docs/storage-classes.

[Reference] https://cloud.google.com/billing/docs/how-to/manage-billing-account

Option 4 is the correct answer. You can run the command `gcloud components install app-engine-python` to install the gcloud components that include the AppEngine extension for Python 3. gcloud components are commands used to perform actions such as listing, installing, updating, or removing Google Cloud SDK components. You can use "install" to install the specified component and check the installation status. App Engine extensions are required to deploy Python applications on App Engine.

Option 1 is incorrect. This is not a valid command. You need to use gcloud components.

Option 2 is incorrect. This is not a valid command. You need to use gcloud components.

Option 3 is incorrect. gcloud components list app-engine-python can list components. This is just a display, so use install to see if the installation was successful.

【reference】 https://cloud.google.com/appengine/docs/standard/python/setting-up-environment

Option 3 is the correct answer. gsutil is a command line for accessing and operating cloud storage from the command line. mb is the command to create a bucket. Create a bucket by typing gsutil mb. Cloud Storage has only one namespace, so you can't create a bucket with a name that another user is already using. Run it with a unique gsutil mb bucket name.

Option 1 is incorrect. Create a bucket with the mb command instead of the gsutil create bucket.

Option 2 is incorrect. You must run gsutil mb "unique bucket name" instead of gsutil mb bucket.

Option 4 is incorrect. Create a bucket with the mb command instead of the gsutil mk bucket.

【reference】 https://cloud.google.com/storage/docs/gsutil/commands/mb

Option 3 is the correct answer. If you want to increase the persistent disk space, resize the disk and resize the single file system instead of repartitioning and formatting the disk. Adding a persistent disk volume size to suit your performance needs does not immediately take advantage of the additional space. After that, you will need to extend the file system on disk.

Resizing the disk may require resizing the file system and partitions.

Boot disk: For VMs that use public images, resizing the VM's boot disk and restarting the VM automatically resizes the root partition and file system. If you are using an image that does not support this feature, you will need to manually resize the root partition and file system.

Non-boot disk: When you resize the disk, you need to expand the file system on the disk to use the added space.

Option 1 is incorrect. This is the response when a new disc is attached. After creating a new disk and attaching it to the VM, you need to format and mount the disk. This will allow the operating system to use the storage capacity.

Option 2 is incorrect. If you want to increase the persistent disk space, resize the disk and resize a single file system instead of repartitioning and formatting the disk.

Option 4 is incorrect. The gcloud disk resize command is not a valid command. for example, If you are using ext4, Use the resize2fs command to extend the file system.

Option 3 is the correct answer. You can create an instance template by running the command `gcloud compute instance-templates create [new template name]`. The gcloud command-line tool allows you to manage Compute Engine resources using the gcloud compute command group. After that, by writing instance-templates create, it will create instance-templates of the Compute Engine resource.

Option 1 is incorrect. The template is incorrectly written in "gcloud compute template create". Enter instance-templates instead of template.

Option 2 is incorrect. Enter the gcloud compute command group instead of gcloud create compute instance-templates.

Option 4 is incorrect. Enter the gcloud compute command group instead of gcloud instance-group instance-templates create.

【reference】 https://cloud.google.com/sdk/gcloud/reference/compute/instance-templates/create

Option 1 is the correct answer. App Engine is a fully managed serverless platform for developing and hosting large-scale web applications. From your application, you can choose from multiple popular programming languages, libraries, and frameworks, and App Engine provides on-demand server provisioning and scaling for your application instances.

Option 2 is incorrect. Compute Engine is not a fully managed service, so you have to manage the infrastructure yourself, which has more management overhead than App Engine.

Option 3 is incorrect. Kubernetes Engine requires you to set up and manage the Kubernetes environment yourself, and requires skills and settings related to Kubernetes compared to App Engine.

Option 4 is incorrect. Cloud Functions are typically used to respond to Google Cloud events, not to run continuously running applications.

【reference】 https://cloud.google.com/appengine/docs/standard

Option 1 is the correct answer. Use "Alignment" to group a set of metrics that arrive over a period of time into a regular size bucket. Alignment is one of the processes used when setting metrics for monitoring using the operation suite. You can convert data points to a single data point and group them.

This process takes all the data points within the Alignment period, applies mathematical transformations such as mean, minimum, maximum, and delta, and transforms them into a single data point for each period. This allows you to group a set of metrics that arrive over a period of time into a regular size bucket.

Option 2 is incorrect. Aggregation is the entire process for combining data, which includes Alignment. Time series aggregation is done in two steps. First, each time series in the set is aligned to the same time interval boundary, then the number of time series sets is optionally reduced.

Options 3 and 4 are incorrect. There is no such process.

【reference】https://cloud.google.com/monitoring/api/ref_v3/rest/v3/projects.alertPolicies#aggregation