The ClusterView should display the hardware as exactly as possible.
An HDD is subdivided into clusters that correspond mostly to the size of pages in Windows. These are arranged approximately as they appear in the classic ClusterView.
This basic view is, however, too imprecise for SSDs. SSDs are also subdivided into clusters but they manage these clusters in blocks. This changes the type of allocation as well as the consequences of fragmentation.
If an HDD becomes fragmented, the speed and wear of the HDD are relatively unaffected by the fragments being combined in one group or arranged consecutively in a row. Small gaps don’t pose any problem although the fact remains that data arranged consecutively can still be read faster as a result of the read/write head‘s movements being kept to a minimum.
With SSDs it’s entirely different. When fragments are combined, there may be more blocks occupied than are actually needed. That’s why it’s important to combine all fragments and have data written consecutively so that any gaps can be removed. Doing so can potentially reduce the number of occupied blocks and thereby extend the lifespan and performance of the SSD considerably.
The new ClusterView displays the blocks on an SSD along with any clusters they contain. In extreme cases, a clear reduction in the number of occupied blocks will be visible. It’s also possible to read if there are any poorly occupied blocks present. There may, for example, be one block with two occupied clusters and another with three occupied clusters. These two blocks could then be combined.