Expand my Community achievements bar.

SOLVED

Recommended distributed file system for a shared datastore author cluster

Avatar

Level 2

We run CQ 5.6.1 on Redhat Linux and would like to switch to a shared datastore author cluster.

The question is: how to best sync the datastore folder in the author cluster?

E.g. is there any recommended or tested approach by Adobe, such as 

http://en.wikipedia.org/wiki/Distributed_Replicated_Block_Device

or any of the ones mentioned here: http://en.wikipedia.org/wiki/Comparison_of_distributed_file_systems (e.g. Lustre)

1 Accepted Solution

Avatar

Correct answer by
Employee Advisor

Hi,

This question has 2 answers:

1) From a support perspective please refer to the technical requirements documentation [1]; there the filesystems (and their underlying block layer technology like LVM or DRBD) are not listed. If you want to be 100% sure, please ask support.

2) From an consultant's experience the underlying block layer doesn't really matter, as long as you have experience in operating and troubleshooting it. Also the filesystem doesn't really matter, as long as it can store lot of potentially large files.

kind regards,
Jörg

[1] http://docs.adobe.com/docs/en/cq/5-6-1/deploying/technical_requirements.html

View solution in original post

4 Replies

Avatar

Correct answer by
Employee Advisor

Hi,

This question has 2 answers:

1) From a support perspective please refer to the technical requirements documentation [1]; there the filesystems (and their underlying block layer technology like LVM or DRBD) are not listed. If you want to be 100% sure, please ask support.

2) From an consultant's experience the underlying block layer doesn't really matter, as long as you have experience in operating and troubleshooting it. Also the filesystem doesn't really matter, as long as it can store lot of potentially large files.

kind regards,
Jörg

[1] http://docs.adobe.com/docs/en/cq/5-6-1/deploying/technical_requirements.html

Avatar

Level 2

Not really. I already knew those docs, but they only say:

"The parameter path points to the location on a shared file system where the data store will be held."

But which file system do I use? What is recommended? NFS is known to be slow, so any other recommended options?

Or better: does someone run it this way and if yes what do you use?

Avatar

Level 10

I have passed this question to some of our internal consultants. Also - i will ensure we log a ticket against the docs seeking more detailed information on this use case.