Mosaic datasets—the data model for managing raster data in ArcGIS—can be stored in a file or enterprise geodatabase. File geodatabases provide fast access and the required scalability for most applications. The main application for using enterprise geodatabases is for workflows that multiple users may be editing the mosaic dataset simultaneously, as well as scaling to large (more than 100,000 rasters) enterprise or cloud implementations. For these types of workflows, it is generally assumed that file geodatabases are being used, but the process of using enterprise geodatabases is the same.
Location of mosaic datasets
When working on small collections of files connected through a removable drive, you can store the mosaic dataset in a root directory of the data. The advantage of this is that if the drive letter or location of the files change, ArcGIS can still access the imagery, as the system checks for files in a relative path to the mosaic dataset if the imagery cannot be accessed directly. For larger implementations and optimization, as a general rule, it is better to store the mosaic datasets in a separate set of dedicated directories, making it simpler to optimize access to these small files and assist in backing them up. By default, when overviews are created for mosaic datasets, they are stored in the same location as the mosaic datasets. When these overviews are small, this is suitable, but when working with mosaic datasets that have large overviews, it is often advantageous to define the location of the overviews when the overviews are created so that they are stored similarly to the imagery.
File geodatabases provide very good performance, but access through desktop or server can be busy, with large numbers of requests made to the file system. It is recommended that you store the mosaic datasets on a drive to which the server has fast access, and ensure that the file geodatabase is on a faster direct-access drive on the server. This is especially true when working with mosaic datasets that contain hundreds of thousands of records. When a large mosaic dataset is stored in a network file share, it can be detrimental to performance. The reason is that if the files are shared, the operating system needs to check the indices of these files every time they are read, as it is possible that a separate machine has written to the file between any two read requests. Because the file geodatabase is busy, this continual checking of a large number of records reduces performance. One option is to store the mosaic dataset being used for authoring on a network attached storage (NAS) or a storage area network (SAN). Then, before publishing, copy the mosaic dataset to the appropriate servers using a directory name that is the same as the original.
Mosaic dataset naming
When using large numbers of mosaic datasets, it is useful to use a standardized naming convention. The imagery workflows use the following prefixes:
S_xxx—Source mosaic dataset
D_xxx—Derived mosaic dataset
R_xxx—Referenced mosaic dataset
Number of mosaic datasets per geodatabase
A single geodatabase may contain a large number of mosaic datasets. For file geodatabases, there is no advantage to storing multiple mosaic datasets in a single geodatabase. You can use either a separate geodatabase for each mosaic dataset or a single geodatabase for a small group of related mosaic datasets that define a project, making backup and restoring simpler.