@calivm - тут про WB пишут, но я думаю что и к capacity это тоже относится:
Block allocation in the WB region is performed in a round-robin fashion keeping wear leveling in mind. Even when a write operation is an overwrite of data that is already in the WB, VSAN never does an in-place re-write of an existing SSD page. Instead, a new block is allocated and metadata are updated to reflect that the old blocks are invalid. VSAN fills an entire SSD page, before it moves to the next one. Eventually, entire pages are freed when all their data are invalid (it is very rare that VSAN may have to re-buffer data to allow for SSD pages to be freed). Also, because the device firmware does not have visibility into invalidated data, it sees no “holes” in pages. Thus, internal write leveling (by moving data around to fill holes in pages) is practically eliminated. This has a very positive impact on the overall endurance of a device. In general, VSAN’s design has gone to great lengths to impose a workload that is very benign to the endurance of SSDs. As a result, we expect that the life expectancy of SSDs in VSAN may exceed the manufacturers’ specifications, which are developed with more generic workloads in mind.