A set of notebooks to explore and explain core conceptions of Apache Hudi, such as file layouts, file sizing, compaction, clustering and so on.
① The notebooks manipulate a public dataset: amazon-reviews-pds, the location is s3://amazon-reviews-pds, it is accessible on aws global regions, for China regions or non aws users, you can download it to local with S3 client tools.
② The running environment of notebooks is Amazon EMR Studio, a managed notebook service for Amazon EMR. If you have no aws accounts, you can modify notebooks to adapt to a notebook environment which supports Spark kernal.
③ The recommended configuration for Spark cluster is: 32 vCore,120GB or higher, the master node must have 100GB+ free disk space.
Update Notes
@2023-08-22: The public dateset "amazon-reviews-pds" on s3://amazon-reviews-pds is closed recently, you can download raw data from: https://cseweb.ucsd.edu/~jmcauley/datasets/amazon_v2/, but the data format and schema are different with original parquet files on s3://amazon-reviews-pds, you need clean & format raw data by yourself.