I have a good understanding of how disk drives and file systems work. I also have no difficulty understanding the typical single drive defragmentation process. I also understand the various RAID technologies. Put them together and I get a real mental block. Do you have an explanation?
Let's start with a plain hard disk. Files are written in chunks of data. When you are not lucky (e.g. the disk gets full or you have had many deletions and creations of new files) these chunks are not continuously placed on the disk. The more small files you have, the more files might get fragmented - performance suffers. Also, you may face the so-called free space fragmentation, i.e. free space is scattered throughout the disk, thus leading to a higher fragmentation of newly created files, since these are right from the beginning placed in the "holes". This also causes severe performance penalties. That was easy, wasn't it? ;)
Okay, now imagine a RAID 5 system is writing a file. EVERY file will be broken up into pieces and placed on various hard disks including checksums etc. Therefore, every file will be split into smaller pieces. Can you already guess what will occur? BINGO! The free space fragmentation can be a significant impact on your system. And even worse, the fragmentation of all files due to the smaller size of their chunks can increase. But this depends on your system as well as on the usage. So, you actually need a defrag software even more when using RAID compared to regular hard disks. Just give it a try and you will see!
Dig Deeper on Enterprise infrastructure management
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.