Topics within the Google Cloud Professional Data Engineer certification exam

 

Topics within the Google Cloud Professional Data Engineer certification exam

For some years now, I’ve had an hobby within the principles of gadget gaining knowledge of (ML) and wanted to recognise greater. When I came throughout the Google® Cloud Professional Data Engineer certification exam, I become intrigued approximately how ML ideas intertwine with cloud ideas, specially inside the Google Cloud Platform (GCP).

As a Biomedical Engineering pupil at university, I recognise that the principles of ML powerfully relate to how the human brain the whole thing. It’s stimulating to see how computing can use the neuronal influences of the brain. With no history in cloud apart from a yr of revel in within the cloud computing field, I done three Google Cloud certifications—an accomplishment that makes me surely proud.

In this post, I’m sharing the number one sources that I used to observe for this examination and subjects included inside the examination.

Study resources

Here stand the chairs I went to make for taking the examination.

DATA ENGINEERING ON GCP AT FAST LANE

While reading for this certification, I first attended a four-day Google path, which I enormously recommend if you are interested in Big Data and ML inside Google. In this course, I discovered about the gear that GCP affords when ingesting, making ready, and studying Big Data. Some of the important thing subjects were:

·        Challenges confronted with Data Engineering

·        Deep dive into:

·        Big Query

·        Big Table

·        Dataflow

·        Dataproc

·        ML offerings inclusive of Kubeflow

·        Demos and labs of the usage of every carrier

Although this course does now not relate at once to the Professional Data Engineer certification examination, I discovered it quite beneficial and learned quite lots. You can find the direction at Fast Lane Data Engineering on Google Cloud Platform. There are also other guides out of doors of the information engineering music at Fast Lane GCT.

LINUX CONSERVATOIRE

I used Linux Conservatory as my primary source of study cloth. I went thru the entire data engineering direction and observed it very useful. They updated their direction quite lately, so it's miles very unique and gives many beneficial tips and content material that comes up within the examination. Because the practice checks are much like the certification exam, they organized me properly. As a person with little experience inside the facts international, I sense that this direction is ideal in terms of content material and explanation.

COURSERA AND QWIKLABS

Coursera has every other useful direction, that is barely longer than the Linux Academy direction. The Coursera direction has instance eventualities that helped me recollect which equipment are quality for specific customer problems. Also, most sections have Qwiklabs that are pretty useful. They let me use the GCP console table to place principle into fingers-on practice, which improved my expertise.

I focused on the subsequent guides:

·        Data Engineering(DE), Big Data, and Machine Learning on GCP

·        Preparing for the Google Cloud Professional Data Engineer Exam Topics inside the certification examination

The section provides a list of the principle subjects included in the examination, some sub-topics, and on occasion my thoughts about the cloth.

BigQuery (A main cognizance in the exam)

·        Integration with Google Identity and Access Management (IAM) roles

·        Basic knowledge of the GCP Key Management Sevice (KMS) and keys (google-managed, consumer-furnished, and consumer-controlled)

·        Partitioned tables, in particular as utilized in SQL instructions

·        Wildcards

·        Federated tables

·        Integration with Google Cloud Storage (GCS)

·        BigQuery (BQ) information transfer carrier and connectors

·        When to apply normalized and denormalized data

·        Loading one-of-a-kind facts formats into BQ, such as an awesome know-how of the Apache Avro™, CSV, Apache Parquet, and JSON codecs

·        Pricing with slots

·        Cached queries

Dataflow

·        Integration with IAM roles, specially the developer role

·        Differences among global, fixed, consultation, and sliding windows and whilst to apply every kind

·        Best practices on coping with pipeline mistakes, mainly, try-catch-block errors

·        Different varieties of transform techniques, as an instance, Apache Beam ParDo

·        Watermarks

·        Apache Beam

BigTable

·        Schema layout, inclusive of whilst to use tall and slender tables or short and extensive ones

·        Schema that could cause slow performance and how to optimize overall performance

·        When to apply tough disk force (HDD)

·        How to exchange between HDD and solid-country drive (SSD)

Pub/sub

 

·        Process of transferring from the Apache Kafka to pub/sub workflow

·        IAM controls on one-of-a-kind tiers, including the truth that the publisher degree has no IAM controls

·        Learn how the system of message drift works, which include why delays in sending messages would possibly arise

Cloud Spanner

·        Not an awful lot in the exam, simply fundamental principles

·        Primary and secondary indexes

Dataproc

·        Good know-how of the Apache Hadoop atmosphere

·        IAM integration

·        Benefits of preemptible nodes

·        Best practices for migrating Hadoop clusters to Dataproc, together with always isolating statistics from storage by using using GCS

·        Best practices for optimizing overall performance

·        Connectors

·        Apache Spark

Dataprep

·        Not lots inside the examination but you must realize the basic principles

Machine Learning

·        Differences among training and test data.

·        Overfitting and underfitting, along with why they can show up, a way to prevent them.

·        Good expertise of ML sorts, which includes supervised studying, unsupervised gaining knowledge of, reinforcement getting to know, even though I saw no questions about reinforcement.

·        Not an awful lot on Tensorflow, but you have to realize the simple principles.

·        Good know-how of ways neural networks (NN) work. There have been questions about huge NN, deep NN, and both extensive and deep NN.

·        Regularization parameters, together with L1 and L2, inclusive of more than one scenario-based totally questions of when to use every kind.

GCP ML offerings

·        Good understanding of every carrier, particularly Natural Language API, along with sentiment and entity analysis

·        A couple of questions on when it is useful for a purchaser to use ML offerings

·        AI platform, including how it works and on line as opposed to batch predictions

Datalab

·        Basic ideas

·        A query came up approximately how you could percentage notebooks

DataStudio

 

·        Basic standards

·        Caching with BQ, along with query cache and prefetch cache

·        A question came up approximately metrics and dimensions, so you have to know the distinction

Cloud Composer

·        You need to know directed acyclic graph (DAG) documents in detail, which includes the additives.

Extra notes

·        The examination had no case research.

·        The examination didn’t have a lot on Cloud SQL.

·        You need to realize the facts pipelines thoroughly.

·        You should recognize the important thing differences between the information services in GCP.

·        The Google practice examination is also pretty useful, so consider taking it.

Conclusion

I hope the post of my adventure to certification helps you for your journey. Good success to all those who plan to take the examination!

tockhop