Story About AWS Data Analytics Certification

I appeared for the exam on 14th July’23 but got results on 17th August’23 informing me that I have passed with full marks. Literally, I didn’t believe it for a second, not because I am not capable of doing such miracles, but because this is the very first time in my entire lifetime :)

PREPARATION

I had taken up this Udemy course to begin with. Trust me this is not at all enough. I had went through below youtube channels as well for the additional knowledge:

1. Johnny Chivers mostly for the practicals.
2. AWS ReInvents and AWS Techtalks for some of the sessions, like the topics on which I felt I needed more knowledge.
3. Cover all the research papers suggested in the Udemy course. It gives you additional information.
4. AWS services FAQs from the AWS docs.
5. Practice the services from the AWS console, especially those on which you have never worked before. THIS IS VERY IMPORTANT. DON’T SKIP THIS PART.
6. You can try a few workshops for hands-on learning, like this one for the kinesis case study. You can try a few others from this portal for other services in which you need to build confidence.
7. You can also refer to the AWS training material and developer guide.

Be curious. This will help you for sure.

VIDEO AND ARTICLE LINKS TO WHICH I REFERRED

1. Lake Formation
2. Kinesis Case Study
3. Kinesis Workshop
4. Lake Formation Workshop
5. Lake Formation RLS and CLS security
6. Opensearch Serverless
7. AWS ReInvent- Kinesis
8. Cross Account Access S3
9. Redshift
10. DyanmoDB

MOCKS

You can refer to the Udemy tests by Stephane Marek and Bonso. These are not enough though. Try solving as much tests as possible.

EXPERIENCE

Difficulty level is directly proportional to the knowledge you have.

The exam is for 3 hours. Patience is the key. If you can’t sit for a long time, then I would suggest practicing mock tests by taking up one mock test in a single go. This will help you.

Proctors are very strict. Follow all the rules during the exam. As AWS suggests, this should not be your first AWS exam, and you should have at least some experience using these services.

Lots of questions were around Kinesis, DynamoDB, Redshift, Athena, S3, Glue, EMR, Lambda, Opensearch.

I also got couple of questions on AWS MSK. I would also suggest to have knowledge about monitoring applications for example, monitoring FLINK Kinesis Data Analytics application.

Security for the services and Integration of services should be well prepared.

Don’t forget to read about all the services listed in the Syllabus by AWS. At least have some basic knowledge about the services which is not covered in the Stephane’s course.

Access control is again another topic that could be tested during the exam. I received couple of questions on this.

Prepare high availability of the services. Single AZ, multi-AZ and so on. How many nodes under single AZ per service can be created/deployed? What happens in case of failures? Try answering such questions when preparing for the exam. This is what I did and it helped me.

MOST APPEARED KEYWORDS

1. Most performant
2. Cost effective
3. Efficient
4. Highly Available (in rare cases)
5. Minimises Operational Overhead

THE CATCH

There could be some questions where they might trick you. For example, in the question, they might talk about building a real-time pipeline, and as it is known that Kinesis Data Streams is for real-time streaming and Firehose is for near-real time, you might end up choosing KDS as the correct answer, but in fact, the correct answer would be Firehose because the KDS pipeline option mentioned is not end-to-end correct.

For example, KDS can’t integrate directly with S3, but KDF can. So, by seeing real-time in the question, you might choose KDS, but in fact it would be KDF since it integrates directly with S3. This is just an example.

Choose your answers wisely. Highlight the keywords from the question, but think about that.

For example, there was one question where they asked about ETL processing and glue and EMR both being suitable options in this case, but then I looked at the question again and it said that we need to choose a cost-effective solution. Then yes, I had to go ahead with glue since it’s cheaper than EMR.

Another example is that for ad-hoc analytics, it is recommended to use Athena instead of Redshift, but again, it depends on the scenario mentioned in the question. Don’t just fall for the keywords.

CONCLUSION

I feel that I might have overprepared for this exam, but I had to ensure that I didn’t lose the chance. Prepare well, be smart, and rest; everything will be handled.

Again, this is something that is not new. A lot of people appear for this exam and clear it, but the marks here are a special feeling for me. Therefore, it will indeed be a special and memorable one.