Conclusions

---

Although disk optimization is an attractive theory that offers some potential benefit to individual PCs and low-end users, it does not appear to adequately address the complexity resulting from multiple partitions, multiple disks, and multiple users. Consequently, it is unlikely to produce significant performance benefits in an enterprise environment and may sometimes end up costing more in system overhead than it offers in return. Businesses, therefore, should put their faith in proven solutions such as RAID, caching, and defragmentation. If firms deal with reliable vendors, they will realize significant gains in overall network performance. As an enterprise solution, optimization has so many shortcomings that it is not about to replace or enhance network defragmentation any time soon. While it may result in a small gain on occasion under ideal conditions, it is equally likely to slow down a system by instituting arbitrary rules and idealized notions of head movement.

---