5 ms·
It backs its metadata up to S3. You do need metadata to map inodes / slices / chunks to s3 objects, though. Tigris has a one-to-one FUSE that does what you wan
by mrkurt 8mo ago
It backs its metadata up to S3. You do need metadata to map inodes / slices / chunks to s3 objects, though.
Tigris has a one-to-one FUSE that does what you want: https://github.com/tigrisdata/tigrisfs https://github.com/tigrisdata/tigrisfs
- ifoxhz 8mo agoFUSE generally has low overall performance because of an additional data transfer process between the kernel space and user space, which is less than ideal for AI training.
- suavesu 8mo agoThe MLPerf performance of JuiceFS last year, FUSE client, and TCP/IP network. There're some analysis about performance and bottleneck. -> https://juicefs.com/en/blog/engineering/mlperf-storage-v2-ai-training-storage-performance https://juicefs.com/en/blog/engineering/mlperf-storage-v2-ai...