Skip to content

Commit 296e956

Browse files
authored
update setup_gcsfuse.sh (#145)
1 parent c1fc2f5 commit 296e956

1 file changed

Lines changed: 13 additions & 7 deletions

File tree

setup_gcsfuse.sh

Lines changed: 13 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -15,7 +15,10 @@
1515
# limitations under the License.
1616

1717
# Description:
18-
# bash setup_gcsfuse.sh DATASET_GCS_BUCKET=maxdiffusion-github-runner-test-assets MOUNT_PATH=/tmp/gcsfuse
18+
# DATASET_GCS_BUCKET and MOUNT_PATH are required,FILE_PATH is optional, if provided,
19+
# will use ls -R to improve first-time read performance by prefetching metadata
20+
# https://cloud.google.com/storage/docs/cloud-storage-fuse/performance#improve-first-time-reads
21+
# bash setup_gcsfuse.sh DATASET_GCS_BUCKET=maxdiffusion-github-runner-test-assets MOUNT_PATH=/tmp/gcsfuse FILE_PATH=/tmp/gcsfuse/my_dataset
1922

2023
set -e -x
2124

@@ -44,9 +47,12 @@ fi
4447
mkdir -p $MOUNT_PATH
4548

4649
# see https://cloud.google.com/storage/docs/gcsfuse-cli for all configurable options of gcsfuse CLI
47-
# Grain uses _PROCESS_MANAGEMENT_MAX_THREADS = 64 (https://github.com/google/grain/blob/main/grain/_src/python/grain_pool.py)
48-
# Please make sure max-conns-per-host > grain_worker_count * _PROCESS_MANAGEMENT_MAX_THREADS
49-
50-
gcsfuse -o ro --implicit-dirs --http-client-timeout=5s --max-conns-per-host=2000 \
51-
--debug_fuse_errors --debug_fuse --debug_gcs --debug_invariants --debug_mutex \
52-
--log-file=$HOME/gcsfuse.json "$DATASET_GCS_BUCKET" "$MOUNT_PATH"
50+
TIMESTAMP=$(date +%Y%m%d-%H%M)
51+
gcsfuse -o ro --implicit-dirs --log-severity=debug \
52+
--type-cache-max-size-mb=-1 --stat-cache-max-size-mb=-1 --kernel-list-cache-ttl-secs=-1 --metadata-cache-ttl-secs=-1 \
53+
--log-file=$HOME/gcsfuse_$TIMESTAMP.json "$DATASET_GCS_BUCKET" "$MOUNT_PATH"
54+
55+
if [[ ! -z ${FILE_PATH} ]] ; then
56+
FILE_COUNT=$(ls -R $FILE_PATH | wc -l)
57+
echo $FILE_COUNT files found in $FILE_PATH
58+
fi

0 commit comments

Comments
 (0)