Move sas7bdat files off the SAS server

[jawad_at_msc]

Hi,

I’m not sure if this is a silly question, but for security reasons, we need to move all SAS datasets (.sas7bdat) off the server hosting Office Analytics.

We have two servers: Metadata and Office Analytics v9.4, both running on Windows.

Currently, all end users access datasets on the SAS server using their client software Enterprise Guide.

So, we plan to relocate all datasets (SAS tables) from the SAS server to a file server and we are trying to figure out how to do this.

My questions are : Is it considered a bad practice to separate the datasets from SAS OA Server? And what key considerations should we be aware of for this migration?

Thank you for any insights.

[Kurt_Bremser]

Use a file server only for backup or archive. Network shares have severe impacts on performance.

If security is an issue, make sure your SAS server is the secure environment.

[jawad_at_msc]

Hi Kurt,

Thank you for your response.

We want to provide access to the data (sas7bdat files) without using SMB shares, as we aim to prevent users from accessing application servers directly.

Do you have any recommendations, considering that users access SAS through a desktop client only?

Regards

[Kurt_Bremser]

This translates to

"We want our user to use the SAS server without using the SAS server".

Please provide more details, as up to now this sounds just like the stupid idea of a total idiot, aka "Manager".

[SASKiwi]

In most modern data centres, servers no longer have dedicated storage, but share SAN (Storage Area Network) storage. In this scenario, there is NO difference between SAS server and file server storage as they just share the same storage resources. If your architecture is different, please explain.

[Sajid01]

Hello @jawad_at_msc 
This is something that can be easily accomplished. But you will have to involve your Windows Admin/infrastructure/storage teams. 
1.An additional disk can be attached to your server. While executing the SAS codes the output data sets are saved here.
2.Provide access to a shared drive. This is the final location for your datasets.
You can dispense with the local disk and directly use shared drive. But this will result in performance degradation. Read/Write will be slow.