When you format a disk as NTFS in Windows 2000, Windows automatically chooses the optimum cluster or allocation unit size for the disk. The allocation unit size is the smallest amount of disk space that can be allocated to hold a file. Because files are almost never exactly the same size as a multiple of the allocation unit size, every disc has a certain amount of wasted space on it. The amount varies with file size and cluster size.
When a drive is first formatted in NT, the largest cluster size that can be selected by default (i.e., without the user delivering specific override commands) is 4K. The biggest reason for this is that NTFS file compression doesn't work on drives with clusters larger than 4K. In fact, it's not just NTFS compression that may not work, but a good many other disk utilities can malfunction when clusters sizes are set to larger than 4K.
The CONVERT.EXE utility, which is used to convert a FAT partition to NTFS, sets the cluster size to 512 bytes -- same as the byte-boundary size used by FAT partitions. Partitions formatted during NT 4.0 setup use a 512-byte boundary size because they are formatted as FAT and then converted to NTFS. However, Windows 2000 more intelligently chooses cluster sizes based on the size of the formatted partition.
Many third-party utilities, such as Powerquest's PartitionMagic, allow a user to change the cluster size of a partition non-destructively. Since larger cluster sizes do make disk access
Serdar Yegulalp is editor of the Windows Power Users Newsletter. Check it out for the latest advice and musings on the world of Windows network administrators -- and please share your thoughts as well!
Editor's Note: This tip originally appeared on SearchWin2000.com
This was first published in February 2002