Home > Data Protection > PowerProtect DD Series Appliances > Dell PowerProtect Data Domain SISL Scaling Architecture > Redundancy identification and read speed
The problem with finding duplicates exclusively with index lookups is that every disk access only retrieves one segment. One key to disk efficiency is to retrieve many segments with each access. Generally, a given small segment of data in most backups will tend to be stored sequentially with the same neighboring segments before it and after it most of the time. The Data Domain system stores these neighbors together as sequences of segments in units called segment localities, which are packed into containers. The Data Domain file system is a log structured system, at the heart of which is the log of containers storing localities. A locality keeps segments close together on disk when they are neighbors. The system can access all the fingerprints or a whole locality with a single disk access. This means many related segments or segment fingerprints can be accessed very efficiently.