Opinie (ETL Processing on Google Cloud Using Dataflow and BigQuery (Python))

Joshua W. · Sprawdzono prawie 7 lat temu

Kunal L. · Sprawdzono prawie 7 lat temu

Joshua W. · Sprawdzono prawie 7 lat temu

Steve T. · Sprawdzono prawie 7 lat temu

Bernd S. · Sprawdzono prawie 7 lat temu

Michael I. · Sprawdzono prawie 7 lat temu

Mark C. · Sprawdzono prawie 7 lat temu

Alexandre M. · Sprawdzono prawie 7 lat temu

shit lab. full of bugs. did it 3 times and cant ever complete. slow spinning up and down of workers. atrocious. and such limited time. and there is issue recognising tasks i have already completed.

Decki K. · Sprawdzono prawie 7 lat temu

Oren L. · Sprawdzono prawie 7 lat temu

Ekta R. · Sprawdzono prawie 7 lat temu

Richard C. · Sprawdzono prawie 7 lat temu

Google Cloud Shell not working.

Narendra Babu O. · Sprawdzono prawie 7 lat temu

嘉瑋 姚. · Sprawdzono prawie 7 lat temu

good

Pooja T. · Sprawdzono prawie 7 lat temu

Xiangyang S. · Sprawdzono prawie 7 lat temu

Muhammad F. · Sprawdzono prawie 7 lat temu

This lab throws around terminology via the python script names like transformation, enrichment, and mart, but really things like transformation and enrichment would be separate steps from ingestion and not combined into one step. Also, some of the documentation in the lab is erroneous. For example in Step 6, it says in the step you will "Ingest the files from GCS" but that is not true -- Steps 6 uses data only from BigQuery and not from GCS. Furthermore, when you look at the resulting data from the last step, the process has replaced things like City with the customer name, so the example code is flawed. Finally, there is no architecture or data flow diagram -- no visual representation of what is going on; and also no mention of what Apache Beam is and how it relates to DataFlow. Finally, ETL is an outdated legacy term. Big Data uses other concepts and follows a different model than ETL.

Luther M. · Sprawdzono prawie 7 lat temu

Lab worked but some of the steps take an incredible amount of time if you sit and wait for them to complete on the CLI. Also "read the code" isn't a great explanation of each step. Each step feels like it was simply cut/pasted 5 times.

Jamal M. · Sprawdzono prawie 7 lat temu

Benjamin R. · Sprawdzono prawie 7 lat temu

Debanjan D. · Sprawdzono prawie 7 lat temu

Balaji C. · Sprawdzono prawie 7 lat temu

Matthew W. · Sprawdzono prawie 7 lat temu

Anil Sagar R. · Sprawdzono prawie 7 lat temu

Rendson F. · Sprawdzono prawie 7 lat temu