GCP Cert

GCP Certifications

I've worked at Google for 3+ years as a cloud engineer. Although there are internal classes and labs available at Google (also available to outsiders for pay), there is no advantage or special track internally. We have to pass our certifications just like the rest of the world.

For me, I have to study, work through labs, and then find some pet projects on my own to really learn and comprehend it.

/* The following is all from 2019. Some of the items are dated, such as the Case Studies, which have changed a little. But it's still a good guide. */

Basically there are three (3) Google Cloud certification levels:

Associate – Cloud Engineer
Professional
1. Cloud Architect // This is what I took
2. Cloud Developer
3. Data Engineer
4. Cloud DevOps Engineer
5. Cloud Security Engineer
6. Cloud Network Engineer
7. Collaboration Engineer
User certification

I passed the Cloud Architect certification and will mostly focuse on that, which assumes you have 3+ years industry experience, including 1+ years on Google Cloud. I did not have the Google Cloud experience, but took the time to learn and apply new information to my base knowledge.

What I Did...

Sign up for the exam

What it is: https://cloud.google.com/certification/cloud-architect
Register for Exam: https://www.webassessor.com/googlecloud/
- I picked: Google Cloud Certified – Professional Cloud Architect (English)
- Pay $200

Guides and My Notes

The percent is the rough estimate of time I spent using the resource, and what I feel contributed to my success.

30% – Coursera online courses
1. Preparing for the Google Cloud Professional Cloud Architect Exam
2. Complete “Architecting with Google Cloud Platform” course, which covers six areas (select each one individually). Some have changed since I took my exam.
  - Google Cloud Platform Fundamentals: Core Infrastructure – 3 hours
  - Essential Cloud Infrastructure: Foundation – 2 hours
  - Essential Cloud Infrastructure: Core Services – 2 hours
  - Elastic Cloud Infrastructure: Scaling and Automation – 2 hours
  - Elastic Cloud Infrastructure: Containers and Services – 1 hour
  - Reliable Cloud Infrastructure: Design and Process – 5 hours
20% – Instructor Led Course, Architecting with GCP: Infrastructure class
20% – Experience – at Google and previously on my own
1. Training/Qwiklabs – especially “Cloud Architecture” quest (Specifically helps with prep for cert)
  - Cloud IAM: Qwik Start – approx. 30 mins
  - Set up Network and HTTP Load Balancers – approx. 40 mins
  - Networking 101 – approx. 75 mins
  - Networking 102 – approx. 75 mins
  - Getting Started with Cloud KMS – approx. 30 min
  - Orchestrating the Cloud with Kubernetes – approx. 75 mins
  - Stackdriver: Qwik Start – approx. 40 mins
  - Application Performance Management (APM) with Stackdriver – approx. 45 mins
  - Deployment Manager – Full Production – approx. 90 mins
  - Continuous Delivery Pipelines with Spinnaker and Kubernetes Engine – approx. 55 mins
2. Presenting to customers as a Googler
3. GCP experimenting and “playing” on my own $300 credit on my personal Gmail account.
4. Past experience
20% – Notes from other People
1. Study the case studies (Mountkirk Games, Dress4Win, & TerramEarth). Seemed like 40-50% dealt with these cases.
2. Coursera - This is the only online course I used.
3. Linux Academy – has a really good course around the GCP Certification, though it costs money (around $50/mo) to subscribe. First week is free. Some say that the content is much better than Coursera; however, I don't know this personally.
4. Notes from what’s covered on the exam: https://medium.com/@sathishvj/ notes-from-my-google-cloud-professional-cloud-architect-exam-bbc4299ac30
5. Exam Time
  1. You have 2 hours for the exam of 50 questions. You have a little over 2 minutes per question. I took 1:50 minutes. I tried to keep a calm, steady pace. REMEMBER – It’s multiple choice (4 answers) for all questions.
  2. Answering questions – I started at #1 and went straight through to #50, in order. I answered each questions like this:
    1. If EASY – I answered it and clicked NEXT (and didn’t look back)
    2. If MODERATE – I gave it my best shot AND marked it as a question to review again
    3. If HARD – I left it blank AND marked it as a question to review again
  3. I think your first hunch on moderate questions are usually correct. Don’t over psych yourself out.
  4. When confronted with a moderate or difficult question, try to eliminate the obvious wrong answers. Usually there are 2 answers that are obviously not full or completely correct.
  5. Go back and work on the 10 or so difficult questions at the end… but do “guess” something.
10% – Practice Exams
1. Professional Cloud Architect Exam Guide – cloud.google.com/certification/guides/professional-cloud-architect/
2. Professional Cloud Architect Exam Readiness Test – cloud.google.com/certification/practice-exam/cloud-architect
3. Whizlabs provides a set of practice exams for the GCP Certification: whizlabs.com/my-account/
4. ROI Training
  - Quiz 1 – bit.ly/ROI-quiz1
  - Quiz 2 – bit.ly/ROI-quiz2
  - Quiz 3 – bit.ly/ROI-quiz3
  - Quiz 4 – bit.ly/ROI-quiz4
  - Quiz 5 – bit.ly/ROI-quiz5

My Notes of Stupid Thing I Keep Forgetting

weidongzhou.wordpress.com/2017/06/10/google-cloud-sql-vs-cloud-datastore-vs-bigtable-vs-bigquery-vs-spanner/
weidongzhou.files.wordpress.com/2017/06/gcp_storage_option_tree.jpg
Cloud BigTable – time series
Persistent disk performance is based on the total persistent disk capacity attached to an instance and the number of vCPUs that the instance has. Incrementing the persistent disk capacity will increment its throughput and IOPS, which in turn improve the performance of MySQL
Deploy the applications to App Engine Standard and store the data in Cloud Datastore

Things I Should Have Studied More

Load Balancing, especially what is Layer 4 (http) and Layer 7 (https) and the business drivers. https://cloud.google.com/load-balancing/
Deployment and CI/DI
If in doubt use Cloud Storage and BigQuery. Seemed like that was a common theme… but I could be wrong.