Ceph or Hadoop?

Question

Data storage

Hello, please tell me what is better to use ceph or hadup for storing files. We will store small files - ordinary office documents, maybe large files, but rarely. We want to implement a web client and an internal application for viewing, downloading and uploading. Also on Hadup it might be possible to do some kind of distributed computing? We have a university and if file storage can be combined with some kind of science, it would be nice. The main goal is, of course, a distributed file system.

Answer 1

Add NFS and CIFS to the list. Classic file washer.

Answer 2

I have not worked with both, but, as far as I understand, Hadoop is certainly not suitable, unless it is about offline storage and processing of documents.
It looks like you're left with Ceph, which by the way implements the S3 interface.
But MongoDB has GridFS and it scales well. GridFS splits files and stores them in chunks of 255 kB.
There are also GlusterFS and other solutions.