Dataflow pipelines can also run on alternate runtimes like Spark and Flink, as they are built using the Apache Beam SDKs Reference: https://cloud.google.com/dataflow/

By default, prefer field promotion. Field promotion avoids hotspotting in almost all cases, and it tends to make it easier to design a row key that facilitates queries.
Reference: https://cloud.google.com/bigtable/docs/schema-design-time-
series#ensure_that_your_row_key_avoids_hotspotting

Dataproc's Compatibility with Apache Spark: Dataproc is a managed service for running Hadoop and Spark clusters on Google Cloud. This means it is designed to seamlessly run Apache Spark jobs with minimal code changes. Your existing Spark jobs should run on Dataproc with little to no modification.
Cloud Storage as a Scalable Data Lake: Cloud Storage provides a highly scalable and durable storage solution for your data. It's designed to handle large volumes of data that Spark jobs typically process.
Minimizing Operational Overhead: By using Dataproc, you eliminate the need to manage and maintain a Hadoop cluster yourself. Google Cloud handles the infrastructure, allowing you to focus on your data processing tasks.
Tight Timeline and Minimal Code Changes: This option directly addresses the requirements of the question. It offers a quick and easy way to migrate your Spark jobs to Google Cloud with minimal disruption to your existing codebase.
Why other options are not suitable:
A . Copy your data to Compute Engine disks. Manage and run your jobs directly on those instances: This option requires you to manage the underlying infrastructure yourself, which contradicts the requirement of using managed services.
C . Move your data to BigQuery. Convert your Spark scripts to a SQL-based processing approach: While BigQuery is a powerful data warehouse, converting Spark scripts to SQL would require substantial code changes and might not be feasible within a tight timeline.
D . Rewrite your jobs in Apache Beam. Run your jobs in Dataflow: Rewriting jobs in Apache Beam would be a significant undertaking and not suitable for a quick migration with minimal code changes.

Explanation/Reference:

The columns can be grouped into two types-categorical and continuous columns:
A column is called categorical if its value can only be one of the categories in a finite set. For example, the native country of a person (U.S., India, Japan, etc.) or the education level (high school, college, etc.) are categorical columns.
A column is called continuous if its value can be any numerical value in a continuous range. For example, the capital gain of a person (e.g. $14,084) is a continuous column.
Year of birth and income are continuous columns. Country is a categorical column.
You could use bucketization to turn year of birth and/or income into categorical features, but the raw columns are continuous.
Reference: https://www.tensorflow.org/tutorials/wide#reading_the_census_data