Unfortunately there is no short and easy answer to this question because usually one cannot know where a file came from and what it is used for. Generally temporary files and cache files of the internet browsers can nearly always be deleted safely. Besides deleting files, the File Search of the software offers you to move files to a different drive or to a compressed ZIP file (preserving the file system hierarchy) which can then be archived. That way files that are still needed can be easily restored, either manually or by using the restore script that the software is able to generate optionally.
You can use SpaceObServer in a clustered server environment by installing it on one of the nodes. That means that a node fail over is not handled, but scheduled scans will be started automatically after the node is running again. You can also install SpaceObServer on each node. If you do so, you have to change the observing server for the scanned roots to each of the nodes, or to none. You can configure the observing server in the edit field "Observing Server" in the "Expert Settings" tab of the "Configure Scans" dialog ("Scan > Configure Scans"). Leave the observing service blank to allow any SpaceObServer service to execute scheduled scans. You can enter multiple servers, separated by semicolon (";"). By default, local drives are stored in the database in their UNC convention (e.g.: "C:\" -> "\\<servername>\C$\"). To ensure that each node is able to scan these drives you have to scan "\\LOCALHOST\C$\" instead. For each node you install SpaceObServer on you need one licence (See General / Registration).