Glusterfs Gett IOPS Caching Attribute

I am creating a GlusterFS share for a heavy Get Attribute workload. This cluster is configured in Replication mode and must contain several TB of data.

In this case, most of the request from clients is of type " Get Attribute " (for example, the last modified Apache).

I am wondering about IO (disk) performance:

• How does GlusterFS perform disk I / O for Get Attribute?
• Does each Get Attribute request provide a basic disk or are they served from the server without affecting disk IO?
• Is the cache size effective in this scenario?
• Is there a request to the replica server every time the Get Attribute reaches the Gluster server?
• Given that Gluster is the preferred server solution, is it better to use the GlusterFS client of the NFS client under this particular workload?

Thanks in advance.