Re:Ask HN: What does your ML pipeline look like?
by htamas on Sunday April 21, 2019 @03:35PM
(#19714070)
Excuse my ignorance, but why is an NFS better than S3? Both are loading from disk to memory of the Tensorflow Serving container, aren't they?
[ Parent
| on HN ]
Re:Ask HN: What does your ML pipeline look like?
by rpedela on Sunday April 21, 2019 @04:02PM
(#19714220)
NFS is faster and it looks like a normal filesystem to the app so you don't need any special file I/O code.