Understanding functional and technical aspects of Google Professional Data Engineer Exam Operationalizing machine learning models

The following will be discussed here:

• Leveraging pre-built ML models as a service
• Impact of dependencies of machine learning models
• Deploying an ML pipeline
• Customizing ML APIs (e.g., AutoML Vision, Auto ML text)
• Continuous evaluation
• Use of edge compute
• Measuring, monitoring, and troubleshooting machine learning models
• Operationalizing machine learning models
• Machine learning terminology (e.g., features, labels, models, regression, classification, recommendation, supervised and unsupervised learning, evaluation metrics)
• Common sources of error (e.g., assumptions about data)
• Hardware accelerators (e.g., GPU, TPU)
• Retraining of machine learning models (Cloud Machine Learning Engine, BigQuery ML, Kubeflow, Spark ML)
• Choosing the appropriate training and serving infrastructure
• Distributed vs. single machine
• ML APIs (e.g., Vision API, Speech API)
• Ingesting appropriate data
• Conversational experiences (e.g., Dialogflow)

Understanding functional and technical aspects of Google Professional Data Engineer Exam Designing data processing systems

The following will be discussed here:

• Architecture options (e.g., message brokers, message queues, middleware, service-oriented architecture, serverless functions)
• Selecting the appropriate storage technologies
• Distributed systems
• Use of distributed systems
• Choice of infrastructure
• Data modeling
• Schema design
• Designing data processing systems
• Capacity planning
• Hybrid cloud and edge computing
• System availability and fault tolerance
• Mapping storage systems to business requirements
• Designing data pipelines
• Job automation and orchestration (e.g., Cloud Composer)
• Tradeoffs involving latency, throughput, transactions
• At least once, in-order, and exactly once, etc., event processing
• Batch and streaming data (e.g., Cloud Dataflow, Cloud Dataproc, Apache Beam, Apache Spark and Hadoop ecosystem, Cloud Pub/Sub, Apache Kafka)
• Data publishing and visualization (e.g., BigQuery)
• Online (interactive) vs. batch predictions

Training Courses Recommended for the Exam Preparation

Training courses are meant to help candidates to learn about the Google exam syllabus and prepare well. It has hands-on labs and expert support that will allow you to get in-depth knowledge of each domain covered in the test. So, these are some of the best training courses offered by Google for the Professional Data Engineer certification exam.

Reference: https://cloud.google.com/certification/data-engineer

Exam Topics

The syllabus of the Google Professional Data Engineer exam is divided into 4 topics, each covering specific knowledge and skills that the candidates need to develop while preparing for the test. A full outline of the exam content can be viewed on the official website. The highlights of the domains covered in the test are as follows:

Topic 1. Designing Data Processing Systems

To answer the questions related to this first topic of the certification exam, the individuals need to demonstrate their proficiency in selecting the proper storage technologies. This includes their understanding of data modeling, schema design, distributed systems, as well as tradeoffs involving throughput, latency, and transactions. Moreover, the applicants need to have the ability to map storage systems to the business needs. It also measures one’s skills in designing data pipelines, designing a data processing solution, as well as migrating data warehousing & data processing.