Ready for CeBIT: Fujitsu, Maxtor and Quantum's latest High-Performance IDE Drives

Reliability Features

Succeeding in the storage market with IDE drives is not particularly easy. The profit margins are small, as customers want fast and reliable drives, while everybody has to save costs. Manufacturers are doing some tough splits, as they have to keep improving their drives in terms of performance and capacity without sacrificing reliability. A broken motherboard, memory or graphics card can be exchanged, but your data usually dies with the hard drive!

The PCB of a hard drive has to be as small as possible in order to keep production costs low.

Three numbers are important for everybody who keeps using the drive intensely. The MTBF (Mean time between failure) has been the most important statement for a long time. This life time specification is of course quite theoretical and not applicable for most home users, as most of us do not keep the computer running 24 hours a day. If you want to get a server hard drive, I would have a quick look at this number.

All other users should primarily focus on the number of start-stop cycles, which is more important for home or office computers. As you can see on the following table, you cannot always get all the information you want. Unfortunately most manufacturers do not publish all the data we would like to see, even though the specs are always inflated with lots of numbers.

Another information is the component lifetime. In my opinion, this number is the only straightforward information you can get. 500,000 hours between failure are ~57 years of permanent operation - quite a lot for a mechanical device, don't you think? 5 years component lifetime sounds very realistic, as most of you might have experienced hard drive head-crashes or other problems within that time already. The number of start-stop cycles is not by far as useful, as you would have to start your computer almost 30 times a day to exceed the component lifetime.