V Luong created ARROW-6796:
------------------------------
Summary: Certain moderately-sized (~100MB)
default-Snappy-compressed Parquet files take enormous memory and long time to
load by pyarrow.parquet.read_table
Key: ARROW-6796
URL: https://issues.apache.org/jira/browse/ARROW-6796
Project: Apache Arrow
Issue Type: Bug
Affects Versions: 0.14.1
Reporter: V Luong
My Spark workloads produce small-to-moderately-sized Parquet files with typical
on-disk sizes in the order of 100-300MB, and I use PyArrow to process these
files further.
Surprisingly, I find that similarly-sized Parquet files sometimes take very
extremely different amounts of memory and time to load using
pyarrow.parquet.read_table. For illustration, I've uploaded 2 such parquet
files to s3://public-parquet-test-data/fast.snappy.parquet and
s3://public-parquet-test-data/slow.snappy.parquet.
Both files have about 1.2 million rows and 450 columns and occupy 100-120MB on
disk. But when they are loaded by read_table:
* `fast.snappy.parquet` takes 10-15GB of memory and 5-8s to load
* `slow.snappy.parquet` takes up to 300GB (!!) of memory and 45-60s to load
Since I have been using the default Snappy compression in all my Spark jobs, it
is unlikely that the files differ in the their compression levels. That the
on-disk sizes are similar suggest that they are similarly compressed. So the
fact that `slow.snappy.parquet` takes 10-20x amounts of resources to read is
very surprising.
My benchmarking code snippet is below. I'd appreciate your help to troubleshoot
this matter.
```{python}
from pyarrow.parquet import read_metadata, read_table
from time import time
from tqdm import tqdm
FAST_PARQUET_TMP_PATH = '/tmp/fast.parquet'
SLOW_PARQUET_TMP_PATH = '/tmp/slow.parquet'
fast_parquet_metadata = read_metadata(FAST_PARQUET_TMP_PATH)
print('Fast Parquet Metadata: {}\n'.format(fast_parquet_metadata))
durations = []
for _ in tqdm(range(3)):
tic = time()
tbl = read_table(
source=FAST_PARQUET_TMP_PATH,
columns=None,
use_threads=True,
metadata=None,
use_pandas_metadata=False,
memory_map=False,
filesystem=None,
filters=None)
toc = time()
durations.append(toc-tic)
print('Fast Parquet READ_TABLE(...) Durations: {}\n'
.format(', '.join('{:.0f}s'.format(duration) for duration in durations)))
slow_parquet_metadata = read_metadata(SLOW_PARQUET_TMP_PATH)
print('Slow Parquet Metadata: {}\n'.format(slow_parquet_metadata))
durations = []
for _ in tqdm(range(3)):
tic = time()
tbl = read_table(
source=SLOW_PARQUET_TMP_PATH,
columns=None,
use_threads=True,
metadata=None,
use_pandas_metadata=False,
memory_map=False,
filesystem=None,
filters=None)
toc = time()
durations.append(toc - tic)
print('Slow Parquet READ_TABLE(...) Durations: {}\n'
.format(', '.join('{:.0f}s'.format(duration) for duration in durations)))
```
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
