The move into the digital age has meant more and more critical data, images, music and video are being stored on hard drives than ever before. While in most cases that's been a blessing, it also means understanding and maintaining hard drives is something that is becoming increasingly important.
A hard drive consists of a series of one or more thin metallic platters on which data is written and read by a head which travels across and up and down these platters at breakneck speeds, generally equivalent to the “redline” in most automobile engines, in excess of 5000 RPM.
One of the key realities of hard drive-based storage is that all hard drives eventually fail. No hard drive used on a daily basis will last forever, and they need to be treated as such.
Search engine giant Google uses hundreds of thousands of consumer-grade hard drives to power its search engine, and in 2007 published a study showing, among other things, that the key times for hard drive failure were when hard drives were new, and after three years, with regular failures occurring on an ongoing and increasing basis after that time.
No amount of special cooling or temperature control made a significant difference in improving the life of drives, other than avoiding truly excessive heat or cold. And drives that received heavier data use were also not significantly more likely to fail than those that did not, and in fact were somewhat less failure prone as time wore on.
The study also showed that drives that had one or more physical errors reported on the drive surface were far more likely to fail completely in the next 60 days, and that S.M.A.R.T. hard drive monitoring (a technology built into all recent hard drives) did not predict a significant percentage of hard drives that would fail prior to physical failure, although it did predict a majority.
Today's giant hard drives also have higher data error rates than their smaller predecessors, and my experience has been that these drives require somewhat more attention as a result.
One of the key things to check is when drives are first installed, when data is loaded to the disk. I have now begun a practice of, after installing the operating system and key programs, running the Windows drive error-checking program (fdisk) that is available under My Computer/Properties for each drive.
I have found that new large drives, when first loaded with data, sometimes tend to have errors that can be corrected quickly without data loss by using this practice. I also now recommend on drives of 300GB or above in desktops, and 120GB or larger in laptops, of running a monthly error check/fix cycle to help maintain data integrity.
Two key questions continue to come up on a regular basis from consumers and businesses when it comes to drives and PC's. First, I continue to recommend systems always be left on, rather than constantly switched on and off. While power savings may be minimal for systems that are switched, power supply failures are much more common in systems that are not left on continuously.
Hard drive defragmentation and other maintenance is also much more easily scheduled and completed overnights when systems are left on and not being asked to do other tasks.
Of course, backing up hard drives is now more important, and less expensive, than ever, with large external hard drives with built-in backup programs costing under $200. Installing one of these will be one of your best PC investments of 2008.
Tom Meek is a computer and media consultant working with businesses and individuals on high-tech needs. Another Day In Cyberville is published weekly in print and online via The Gainesville Voice, a weekly publication of The New York Times Regional Newspaper Group. You can reach Tom Meek at firstname.lastname@example.org.