Making the course more accessible to Math Engineering students - Githubissues

emanueledellavalle / streaming-data-analytics

Apache License 2.0

30 stars 11 forks source link

Making the course more accessible to Math Engineering students #7

Open emanueledellavalle opened 1 year ago

emanueledellavalle commented 1 year ago

Elena Musiari was so kind to share her and some of her colleagues' reviews on my lectures from the perspective of a Math Engineering student.

General Pains

several IT specific terms could be shortly explained when used the first time

the distinction among conceptual, logical, physical, and system level
generic terms such as API and metadata
meaning of 1:1, 1:N, N:1 and N:N relationships

Specific Pains

01_Taming-velocity

slide 22: the example of pub-sub is hard to understand, the subscription language is not introduced
slide 29: the distinction between logical and physical is hard to get

02_EPL

slide 32: the output clause presentation is too short and partial. Without the discussion in the practice (see Q6), it would not be possible to understand it

03_scalable-stream-processing-with-kafka-and-spark

slides on kafka, too fast, too much. The following youtube videos were used to learn individually:
slide on Spark, too fast, too much. Ignored because it was possible given the rule of the exam.
slide 49, hard to get that the tasks in the spark executors are kafka consumers in a consumer group. The extended explanation helped a bit, but it did not clarify the problem completely

04_Spark-Structured-Streaming

too much code makes the lecture hard to understand
the practice helps in sorting out this content

06_ksqlDB

slides 16-17: the concept of pull and push queries is hard to grasp
slides 19, 24, and 25: ok in class, but hard to study at home. This video helped