Why You Should Not Run MinIO ....

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

BoredSysadmin

Not affiliated with Maxell
Mar 2, 2019
1,050
437
83
Not to say this article lacks merit, it isn't, but it's also highly opinionated, with creators beating their drum.
 

Bjorn Smith

Well-Known Member
Sep 3, 2019
876
481
63
49
r00t.dk
I agree - I mean obviously why would you put an object storage layer on top of a SAN? That kind of defeats the purpose of having a scalable object storage that can scale to many nodes - but to test out object storage to evaluate whether or not its worth transitioning your storage to I do not see any issues with it - as long as you only evaluate for functionality - not performance or resiliency.

So - while I agree that building a new infrastructure/application that uses minio on top of a SAN would be bad and defeat the purpose of scaleout object storage, I think its just fine to test it out and if you already have a SAN - by all means use some of that storage.

Its similar to building ZFS on top of a SAN - it will work, but why do it :)
 
Last edited:

gea

Well-Known Member
Dec 31, 2010
3,140
1,182
113
DE
Single binary minIO is a perfect add on for ZFS filer to create a single node S3 Cloud/Internet share from a ZFS filesystem, much faster, more reliable and safer than for example a NextCloud stack with many files and services or SMB. ZFS gives redundancy, bitrot protection and snaps in single node mode.

You can use several ZFS filer for a multi node setup to improve availability with many disks and high capacity/performance per node.

MinIO multi node per FC/iSCSI, NFS, SMB or similar is indeed quite pointless