0

I got a question to understand the mount process of Amazon s3 service suing fuse-s3fs, when i mount amazon s3 service bucket, does it copies whole data to my local system or it will just act as view of amazon s3 bucket?

The reason why i want to know is, let say i got a system with 500 MB storage space, can I mount a bucket with data for around 200 GB? If it just act as view then i should, but if whole data is copied locally then certainly not.

I need guidance in this regard.

Thanks.

Saqlain
  • 17,490
  • 4
  • 27
  • 33

1 Answers1

1

By default (i.e. without using caching option), s3fs does nothing more than provide a mount point for your S3 bucket. There will be no local data copies retained. s3fs does however provide an option to utilize a local drive as a cache, which can provide some i/o performance improvements in cases where they are needed (as direct read/write to S3 bucket via the mount is not the quickest process).

Mike Brant
  • 70,514
  • 10
  • 99
  • 103
  • In casse of caching, when the s3fs will make a 'sync' with the amazon s3 bucket? It cache whole bucket or can also cache part of bucket? – Saqlain Feb 20 '13 at 08:26
  • @Saqlain The cache is unbounded, so it could grow as large as the S3 bucket if each file is actually accessed and loaded into cache. This also means that you can use a cron job or similar to periodically purge older items from the cache without interfering with s3fs operation if you need to try to maintain a specified size. – Mike Brant Feb 20 '13 at 16:42