[ https://issues.apache.org/jira/browse/PARQUET-1622?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17678056#comment-17678056 ]
Gang Wu commented on PARQUET-1622: ---------------------------------- The issue raised by [~mwish] above may also exist in the parquet-mr. cc [~xinlishang]] [~gershinsky] > Add BYTE_STREAM_SPLIT encoding > ------------------------------ > > Key: PARQUET-1622 > URL: https://issues.apache.org/jira/browse/PARQUET-1622 > Project: Parquet > Issue Type: New Feature > Components: parquet-cpp, parquet-format, parquet-mr, parquet-thrift > Reporter: Martin Radev > Assignee: Martin Radev > Priority: Minor > Labels: features, pull-request-available > Fix For: 1.12.0, format-2.8.0 > > Original Estimate: 48h > Remaining Estimate: 48h > > Apache Parquet does not have any encodings suitable for FP data and the > available text compressors (zstd, gzip, etc) do not handle FP data very well. > It is possible to apply a simple data transformation named "stream > splitting". Such could be "byte stream splitting" which creates K streams of > length N where K is the number of bytes in the data type (4 for floats, 8 for > doubles) and N is the number of elements in the sequence. > The transformed data compresses significantly better on average than the > original data and for some cases there is a performance improvement in > compression and decompression speed. > You can read a more detailed report here: > https://drive.google.com/file/d/1wfLQyO2G5nofYFkS7pVbUW0-oJkQqBvv/view -- This message was sent by Atlassian Jira (v8.20.10#820010)