Networked storage [GlusterFS or something simillar]

Discussion in 'Software Stuff' started by Shilelis, Nov 10, 2019.

  1. Shilelis

    Shilelis New Member

    Nov 6, 2019
    Likes Received:

    I'd like to achieve something like this:

    Let's assume that I have these servers connected via network:
    Storage Server 1 - 4 TB
    Storage Server 2 - 4 TB
    Storage Server 3 - 4 TB
    Storage Server 4 - 4 TB

    1. So with something like GlusterFS I could connect these like using LVM to a single volume of 16TB - is that correct?

    2. If say I wanted some redundancy I could connect these using GlusterFS with mirroring and thus have 8 TB volume - is that correct?

    3. Are there any other easy to setup options besides GlusterFS that would allow some sort of redundancy?

    4. Does GlusterFS require some sort of master server or is it a P2P thing and all of those connected servers act as a master and slave at the same time?

    5. While I have some idea to the answers to the questions above, the biggest for me is - can I setup multiple client volumes over that single GlusterFS volume? So that for example:

    GLusterFS volume: 16TB

    so can I create client volumes on top like this?

    client 1: xfs 2TB volume on top of GLusterFS volume
    client 2: NTFS 6TB volume on top of GLusterFS volume
    client 3: BTRFS 3TB volume on top of GLusterFS volume
    client 4: APFS 3TB volume on top of GLusterFS volume

    and say 2TB spare.

    6. If I can somehow achieve what is question 5, can I easily resize any of the client volumes to be larger or smaller as needed?

    7. Can I easily extend the pooled volume by connecting more servers?

    8. In case of mirroring - I have to always have an even number of servers?
Similar Threads: Networked storage
Forum Title Date
Software Stuff Software Defined Storage Jul 30, 2018
Software Stuff video storage Sep 25, 2017
Software Stuff Apr 5, 2016
Software Stuff Proxmox VE - qcow2 to Ceph storage workaround Nov 6, 2015
Software Stuff Distributed block storage. Jul 20, 2015

Share This Page