Page size and erase block size

Although some people say that modern SSD drives do not require any special kind of configuration to maximize their performance/lifespan, it is easy to find discussions about this subject in big internet tech forums (anandtech, stackoverflow). Furthermore, almost every linux distributions has a special “recipe” about how to configure the system and get the best from solid state disks drives.

Proper partition alignment to physical pages is an usual concern. Even when emulation of 512-byte sector is available,
misaligned partition layout implies in overhead over SSD controller. A second concern about partition alignment relates to “write intensification”, since two or more physical blocks might be marked to cleanup due to poor data placement (a immediate result of bad partition alignment).

Wear leveling algorithms reduces the write amplification issue through `flash translation layer’ mechanisms mapping LBA addresses to physical pages on flash media by spreading writes “evenly” over flash memory. Such translation layer does not know how the media is logically structured and partition misalignment could not be overcome by translation mechanisms.