mukund-thakur commented on code in PR #999:
URL: https://github.com/apache/parquet-mr/pull/999#discussion_r995112398
##########
parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetFileReader.java:
##########
@@ -1093,10 +1099,38 @@ private ColumnChunkPageReadStore
internalReadFilteredRowGroup(BlockMetaData bloc
}
}
}
- // actually read all the chunks
+ // Vectored IO up.
+
+ List<FileRange> ranges = new ArrayList<>();
for (ConsecutivePartList consecutiveChunks : allParts) {
- consecutiveChunks.readAll(f, builder);
+ ranges.add(FileRange.createFileRange(consecutiveChunks.offset, (int)
consecutiveChunks.length));
+ }
+ LOG.warn("Doing vectored IO for ranges {}", ranges);
+ f.readVectored(ranges, ByteBuffer::allocate);
Review Comment:
> if `delta1` is small enough, you will merge `range1` and `range2` and do a
single scan.
Yes.
> If range1 and range2 are very large so that the resulting range is even
larger, do you still do a single scan, or do you split the large range into
smaller ranges and tradeoff some seek cost for increased parallelism?
We don't split the ranges. If the ranges are big enough, they won't be
merged.
##########
parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetFileReader.java:
##########
@@ -1093,10 +1099,38 @@ private ColumnChunkPageReadStore
internalReadFilteredRowGroup(BlockMetaData bloc
}
}
}
- // actually read all the chunks
+ // Vectored IO up.
+
+ List<FileRange> ranges = new ArrayList<>();
for (ConsecutivePartList consecutiveChunks : allParts) {
- consecutiveChunks.readAll(f, builder);
+ ranges.add(FileRange.createFileRange(consecutiveChunks.offset, (int)
consecutiveChunks.length));
+ }
+ LOG.warn("Doing vectored IO for ranges {}", ranges);
+ f.readVectored(ranges, ByteBuffer::allocate);
Review Comment:
> if `delta1` is small enough, you will merge `range1` and `range2` and do a
single scan.
Yes.
> If range1 and range2 are very large so that the resulting range is even
larger, do you still do a single scan, or do you split the large range into
smaller ranges and tradeoff some seek cost for increased parallelism?
We don't split the ranges. If the ranges are big enough, they won't be
merged.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
