cloud-foundation-fabric/examples/data-solutions/dp-foundation

Data Platform

This module implements an opinionated Data Platform (DP) Architecture that creates and setup projects (and related resources) to be used to create your DP.

The code is intentionally simple, as it's intended to provide a generic initial setup (Networking, Cloud Storage Buckets, BigQuery datasets, etc.) and then allow easy customizations to complete the implementation of the intended design.

The following diagram is a high-level reference of the resources created and managed here:

A demo pipeline is also part of this example: it can be built and run on top of the foundational infrastructure to verify or test the setup quickly.

Design overview and choices

Despite its simplicity, this stage implements the basics of a design that we've seen working well for various customers.

The approach adapts to different high-level requirements:

• boundaries for each step
• clear and defined actors
• least privilege principle
• rely on service account impersonification

The code in this example doesn't address Organization level configuration (Organization policy, VPC-SC, centralized logs). We expect to address those aspects on stages external to this script.

Project structure

The DP is designed to rely on several projects, one project per data stage. The stages identified are:

• landing
• load
• data lake
• orchestration
• transformation
• exposure

This separation into projects allows adhering the least-privilege principle relying on project-level roles.

The script will create the following projects:

• Landing This project is intended to store data temporarily. Data are pushed to Cloud Storage, BigQuery, or Cloud PubSub. Resource configured with 3-months lifecycle policy.
• Load This project is intended to load data from landing to the data lake. The load is made with minimal to zero transformation logic (mainly cast). This stage can anonymization/tokenization Personally Identifiable Information (PII). Alternatively, it can be done in the transformation stage depending on your requirements. The use of Cloud Dataflow templates is recommended.
• Data Lake projects where data are stored. itìs composed of 3 layers that progressively process and define data:
  • L0 - Raw data Structured Data, stored in the adequate format: structured data stored in BigQuery, unstructured data stored on Cloud Storage with additional metadata stored in BigQuery (for example pictures stored in Cloud Storage and analysis of the images for Cloud Vision API stored in BigQuery).
  • L1 - Cleansed, aggregated and standardized data
  • L2 - Curated layer
  • Playground Store temporary tables that Data Analyst may use to perform R&D on data available on other Data Lake layers
• Orchestration This project is intended to host Cloud Composer. Cloud Composer will orchestrate all tasks to move your data on its journey.
• Transformation This project is used to move data between layers of the Data Lake. We strongly suggest relying on BigQuery engine to perform transformations. If BigQuery doesn't have the feature needed to perform your transformation you recommend using Cloud Dataflow together with Cloud Dataflow templates. This stage can optionally be used to anonymiza/tokenize PII.
• Exposure This project is intended to host resources to share your processed data with external systems your data. For the porpuse of this example we leace this project empty. Depending on the access pattern, data can be presented on Cloud SQL, BigQuery, or Bigtable. For BigQuery data, we strongly suggest relying on Authorized views.

Roles

We assign roles on resources at project level setting the appropriate role to groups. We recommend not adding human users directly to the resource-access groups with IAM permissions to access data.

Service accounts

Service Account creation follows the following principles:

• Each service account perform a single task having access to the minimum number of resources (example: the Cloud Dataflow Service Account has access to the Landing project and the Data Lake L0 project)
• Each Service Account has the least privilege on each project.

Service Account Keys

The use of SAK within a data pipeline incurs several security risks, as these credentials, that could be leaked without oversight or control. This example relies on Service Account Impersonation to avoid the creation of private keys.

Groups

We use thress groups based on the required access:

• Data Engineers: the group that handles and runs the Data Hub. The group has Read access to all resources to troubleshoot possible issues with the pipeline. The team also can impersonate all service accounts. Default value: gcp-data-engineers@DOMAIN.COM.
• Data Analyst: the group that performs analysis on the dataset. The group has Read access to the Data Lake L2 project and BigQuery READ/WRITE access to the playground project. Default value: gcp-data-analyst@DOMAIN.COM
• Data Security: the group handling security configurations related to the Data Hub. Default name: gcp-data-security@DOMAIN.com

Virtual Private Cloud (VPC) design

The DP accepts as input an existing Shared-VPC to run resources. You can configure subnets for DP resources specifying the link to the subnet in the network_config variable. You may want to configure a shared-VPC to host your resources if your pipelines may need to reach on-premise resources.

If network_config variable is not provided, the script will create a VPC on each project that requires a VPC: load, transformation, and orchestration projects with the default configuration.

IP ranges, subnetting

To deploy your DP you need the following ranges:

• Load project VPC for Cloud Dataflow workers. Range: '/24'.
• Transformation VPC for Cloud Dataflow workers. Range: '/24'.
• Orchestration VPC for Cloud Composer:
  • Cloud SQL. Range: '/24'
  • GKE Master. Range: '/28'
  • Web Server: Range: '/28'
  • Secondary IP ranges. Pods range: '/22', Services range: '/24'

Resource naming convention

Resources in the script use the following acronyms:

• lnd for landing
• lod for load
• orc for orchestration
• trf for transformation
• dtl for Data Lake
• cmn for common
• plg for playground
• 2 letters acronym for GCP products, example: bq for BigQuery, df for Cloud Dataflow, ...

Resources follow the naming convention described below.

Projects:

PREFIX-LAYER

Services:

PREFIX-LAYER[2]-GCP_PRODUCT[2]-COUNTER

Service Accounts:

PREFIX-LAYER[2]-GCP_PRODUCT[2]-COUNTER

Encryption

We suggest a centralized approach to Keys management, to let the Security team be the only team that can access encryption material. Keyrings and Keys belong to a project external to the DP.

To configure the use of Cloud Key Management on resources you have to specify the key URL on the 'service_encryption_keys'. Keys location should match the resource location. Example:

service_encryption_keys = {
    bq       = "KEY_URL_MULTIREGIONAL"
    composer = "KEY_URL_REGIONAL"
    dataflow = "KEY_URL_REGIONAL"
    storage  = "KEY_URL_MULTIREGIONAL"
    pubsub   = "KEY_URL_MULTIREGIONAL"

We consider this step optional, it depends on customer policy and security best practices.

Data Anonymization

We suggest using Cloud Data Loss Prevention to identify/mask/tokenize your confidential data. Implementing the Data Loss Prevention strategy is out of scope for this example. We enable the service in 2 different projects to implement the data loss prevention strategy. We expect you will use Cloud Data Loss Prevention templates in one of the following ways:

• During the ingestion phase, from Dataflow
• During the transformation phase, from BigQuery or Cloud Dataflow

We implemented a centralized model for Cloud Data Loss Prevention resources. Templates will be stored in the security project:

How to run this script

To deploy this example on your GCP organization, you will need

• a folder or organization where new projects will be created
• a billing account that will be associated with the new projects

The DP is meant to be executed by a Service Account (or a regular user) having this minimal set of permission:

• Org level:
  • "compute.organizations.enableXpnResource"
  • "compute.organizations.disableXpnResource"
  • "compute.subnetworks.setIamPolicy"
• Folder level:
  • "roles/logging.admin"
  • "roles/owner"
  • "roles/resourcemanager.folderAdmin"
  • "roles/resourcemanager.projectCreator"
• Cloud Key Management Keys (if Cloud Key Management keys are configured):
  • "roles/cloudkms.admin" or Permissions: cloudkms.cryptoKeys.getIamPolicy, cloudkms.cryptoKeys.list, cloudkms.cryptoKeys.setIamPolicy
• on the host project for the Shared VPC/s
  • "roles/browser"
  • "roles/compute.viewer"
  • "roles/dns.admin"

Variable configuration

There are three sets of variables you will need to fill in:

prefix             = "PRFX"
project_create = {
  parent             = "folders/123456789012"
  billing_account_id = "111111-222222-333333"
}
organization = {
  domain = "DOMAIN.com"
}

For a more fine grained configuration, check variables on variables.tf and update accordingly to the desired configuration.

Customizations

Create Cloud Key Management keys as part of the DP

To create Cloud Key Management keys in the DP you can uncomment the Cloud Key Management resources configured in the 06-common.tf file and update Cloud Key Management keys pointers on local.service_encryption_keys.* to the local resource created.

Assign roles at BQ Dataset level

To handle multiple groups of data-analysts accessing the same Data Lake layer projects but only to the dataset belonging to a specific group, you may want to assign roles at BigQuery dataset level instead of at project-level. To do this, you need to remove IAM binging at project-level for the data-analysts group and give roles at BigQuery dataset level using the iam variable on bigquery-dataset modules.

Demo pipeline

The application layer is out of scope of this script, but as a demo, it is provided with a Cloud Composer DAG to mode data from the landing area to the DataLake L2 dataset.

Just follow the commands you find in the demo_commands Terraform output, go in the Cloud Composer UI and run the data_pipeline_dag.

Description of commands:

• 01: copy sample data to a landing Cloud Storage bucket impersonating the load service account.
• 02: copy sample data structure definition in the orchestration Cloud Storage bucket impersonating the orchestration service account.
• 03: copy the Cloud Composer DAG to the Cloud Composer Storage bucket impersonating the orchestration service account.
• 04: Open the Cloud Composer Airflow UI and run the imported DAG.
• 05: Run the BigQuery query to see results.

Variables

name	description	type	required	default
organization	Organization details.	object({…})	✓
prefix	Unique prefix used for resource names. Not used for projects if 'project_create' is null.	string	✓
composer_config		object({…})		{…}
data_force_destroy	Flag to set 'force_destroy' on data services like BiguQery or Cloud Storage.	bool		false
groups	Groups.	map(string)		{…}
location_config	Locations where resources will be deployed. Map to configure region and multiregion specs.	object({…})		{…}
network_config	Shared VPC to use. If not null networks will be created in projects.	object({…})		{…}
project_create	Provide values if project creation is needed, uses existing project if null. Parent is in 'folders/nnn' or 'organizations/nnn' format.	object({…})		null
project_id	Project id, references existing project if project_create is null.	object({…})		{…}
project_services	List of core services enabled on all projects.	list(string)		[…]

Outputs

name	description	sensitive
VPC	VPC networks.
bigquery-datasets	BigQuery datasets.
demo_commands	Demo commands.
gcs-buckets	GCS buckets.
kms_keys	Cloud MKS keys.
projects	GCP Projects informations.

TODOs

Features to add in future releases:

• Add support for Column level access on BigQuery
• Add example templates for Data Catalog
• Add example on how to use Cloud Data Loss Prevention
• Add solution to handle Tables, Views, and Authorized Views lifecycle
• Add solution to handle Metadata lifecycle

To Test/Fix

• Composer require "Require OS Login" not enforced
• External Shared-VPC