A company may have multiple EKM servers to improve the transfer of data across remote offices or to provide regional workspaces. For simplicity you may designate one of these servers as “master”, but there is no difference in functionality or installed components between a master and a regional server.
Each EKM server has one default workspace that can be configured for user access, and more workspaces can be created if needed.
Each workspace contains a repository for storing and sharing data. EKM will automatically create a user account for a user when they first sign in to the workspace. Initially the user will have limited permissions, but you can edit their permissions after the account has been created. When a user logs into EKM, they are automatically logged into their default workspace, and can potentially access other workspaces if they have access permission in those workspaces. See Configuring Login-related Access Settings in the Installation Guide for details.
In the example above, the EKM server in New York contains a workspace whose data is shared globally by users in New York, London and Tokyo. Any user who wants to be able to access the shared data on the New York server must have a user account defined in the workspace on the New York server. If a user in London wants to download a file from the server in New York, they must sign in to the global workspace on the New York server. In this case file transfers occur over a Wide Area Network (WAN). File transfer performance in this type of setup may be improved if caching is used. See Caching Data when Transferring Files for more information.
In the sample configuration, the servers in London and Tokyo can have a regional workspace set up that only users in that local region or office can access. This enables those users to collaborate among themselves and share data locally if needed. In this case, user accounts are defined in a workspace on the local EKM server, and file transfers occur over a Local Area Network (LAN).
It is also possible to configure EKM to run jobs in a distributed environment, where data that is centrally stored on the "Master" EKM server is used to run jobs in "Slave" EKM workspaces. This enables globally distributed users to take advantage of remote compute resources in a highly efficient way. See Running Jobs in a Distributed Environment for more information.