From CSP-6166: "After noticing slow performance during first login we did some tracing and found that the application is doing numerous different searches when the user first logs in and accesses the dashboard. The first LDAP search returns all the groups the user belongs to and the subsequent LDAP searches return every user in every single one of the groups from the first search. The impact this has is very large when somebody who belongs to many groups with many users logs in to Confluence. We have extended the period of time that the LDAP data is cached, but this still does not resolve the issue when the users data has not yet be cached. We would like to know what the purpose of the subsequent searches is for and if this can be minimized by changing a property file or making some other modification since the impact on our LDAP server is so big."

I believe that this is afother users as well.