How is a 100 byte file a valid refutation of what you quoted? It quite clearly states that you should leave the cluster size alone if your files are typically small.
well, OK let's take a guess at median filesize 4KB [1]. Then the 4KB cluster wastes some space only for half of the files whereas 16 KB wastes space for bigger proportion plus they waste additional 12KB for 50% of the files.
On the other hand the paper also shows how ridiculous it is to talk about the wasted space - look at figure 14, files smaller than 16KB occupy ~1% of all space. Even if we waste space 4:1 it's still ridiculously small amount of space
"But if your files are generally much larger or tend to grow over time..."
No numbers are given in what you quoted, but when I think about "much larger" I'm thinking of file sizes in the megabyte to gigabyte range, such as a dedicated data drive for a media server or a database server. A 4KB median file size doesn't fit my mental model of "much larger".
