[ https://issues.apache.org/jira/browse/PARQUET-2149?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17541699#comment-17541699 ]
ASF GitHub Bot commented on PARQUET-2149: ----------------------------------------- parthchandra commented on PR #968: URL: https://github.com/apache/parquet-mr/pull/968#issuecomment-1136427603 > thanks., that means you are current with all shipping improvments. the main one extra is to use openFile(), passing in length and requesting randomio. this guarantees ranged GET requests and cuts the initial HEAD probe for existence/size of file. By `openFile()` do you mean `FileSystem.openFileWithOptions(Path,OpenFileParameters)`? While looking I realized the Parquet builds with a [much older version of hadoop](https://github.com/apache/parquet-mr/blob/a2da156b251d13bce1fa81eb95b555da04880bc1/pom.xml#L79) > > > have you benchmarked this change with abfs or google gcs connectors to see what difference it makes there? > > > No I have not. Would love help from anyone in the community with access to these. I only have access to S3. > > that I have. FWIW, with the right tuning of abfs prefetch (4 threads, 128 MB blocks) i can get full FTTH link rate from a remote store; 700 mbit/s . that's to the base station. once you add wifi the bottlenecks move. Wow! That is nearly as fast as local HDD. At this point the bottlenecks in parquet begin to move towards decompression and decoding but IO remains the slowest link in the chain. One thing we get with my PR is that the ParquetFileReader had assumptions built in that all data must be read before downstream can proceed. Some of my changes are related to removing these assumptions and ensuring that downstream processing does not block until an entire column is read so we get efficient pipelining. What does the 128 MB block mean? Is this the amount prefetched for a stream? The read API does not block until the entire block is filled, I presume. With my PR, parquet IO is reading 8MB at a time (default) and downstream is processing 1MB at a time (default) and several such streams (one per column) are in progress at the same time. Hopefully, this read pattern would work with the prefetch. > Implement async IO for Parquet file reader > ------------------------------------------ > > Key: PARQUET-2149 > URL: https://issues.apache.org/jira/browse/PARQUET-2149 > Project: Parquet > Issue Type: Improvement > Components: parquet-mr > Reporter: Parth Chandra > Priority: Major > > ParquetFileReader's implementation has the following flow (simplified) - > - For every column -> Read from storage in 8MB blocks -> Read all > uncompressed pages into output queue > - From output queues -> (downstream ) decompression + decoding > This flow is serialized, which means that downstream threads are blocked > until the data has been read. Because a large part of the time spent is > waiting for data from storage, threads are idle and CPU utilization is really > low. > There is no reason why this cannot be made asynchronous _and_ parallel. So > For Column _i_ -> reading one chunk until end, from storage -> intermediate > output queue -> read one uncompressed page until end -> output queue -> > (downstream ) decompression + decoding > Note that this can be made completely self contained in ParquetFileReader and > downstream implementations like Iceberg and Spark will automatically be able > to take advantage without code change as long as the ParquetFileReader apis > are not changed. > In past work with async io [Drill - async page reader > |https://github.com/apache/drill/blob/master/exec/java-exec/src/main/java/org/apache/drill/exec/store/parquet/columnreaders/AsyncPageReader.java] > , I have seen 2x-3x improvement in reading speed for Parquet files. -- This message was sent by Atlassian Jira (v8.20.7#820007)