[
https://issues.apache.org/jira/browse/ARROW-10226?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17210328#comment-17210328
]
Andy Grove commented on ARROW-10226:
------------------------------------
Just tracking progress with debugging this. The issue is that the projection is
behaving differently PER BATCH within these Parquet files. We expect
l_returnflag to be a single char but sometimes the parquet reader is returning
the contents of the l_comment field instead.
{code:java}
[/mnt/tpch/s1/parquet/lineitem/part-00002-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: N
[/mnt/tpch/s1/parquet/lineitem/part-00000-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: N
[/mnt/tpch/s1/parquet/lineitem/part-00001-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: A
[/mnt/tpch/s1/parquet/lineitem/part-00003-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: R
[/mnt/tpch/s1/parquet/lineitem/part-00002-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: N
[/mnt/tpch/s1/parquet/lineitem/part-00000-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: R
[/mnt/tpch/s1/parquet/lineitem/part-00003-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: R
[/mnt/tpch/s1/parquet/lineitem/part-00002-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: s among the fluffily r
[/mnt/tpch/s1/parquet/lineitem/part-00000-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: eposits a
[/mnt/tpch/s1/parquet/lineitem/part-00001-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: N
[/mnt/tpch/s1/parquet/lineitem/part-00001-36eb4379-93a2-47a8-873a-d0f1ed13a85a-c000.snappy.parquet]
first non-null value for l_returnflag in this batch: y ironic foxes above t
{code}
> [Rust] [DataFusion] TPC-H query 1 no longer completes for 100GB dataset
> -----------------------------------------------------------------------
>
> Key: ARROW-10226
> URL: https://issues.apache.org/jira/browse/ARROW-10226
> Project: Apache Arrow
> Issue Type: Bug
> Components: Rust, Rust - DataFusion
> Reporter: Andy Grove
> Assignee: Andy Grove
> Priority: Major
> Fix For: 2.0.0
>
>
> I re-installed my desktop a few days ago (now using Ubuntu 20.04 LTS) and
> when I try and run the TPC-H benchmark, it never completes and eventually
> uses up all 64 GB RAM.
> I can run Spark against the data set and the query completes in 24 seconds,
> which IIRC is how long it took before.
> It is possible that something is odd on my environment, but it is also
> possible/likely that this is a real bug.
> I am investigating this and will update the Jira once I know more.
> I also went back to old commits that were working for me before and they show
> the same issue so I don't think this is related to a recent code change.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
