Hard Disk Drives are as important as a staple diet. They are involved with everything that happens on your computer, the Internet and pretty much anything you access over a network. Hard Disk Drives like most things in computing and vary widely in type and cost. However they all largely rely on the same basic principle, a circular spinning platter consisting of a substrate coated in a magnetic layer, across either a, or several, magnetic heads fly on an armature to read and write information.
There are many different types of hard disk, ranging from the very first MFM type drives, to IDE or PATA as they now seem to be called, SATA, SCSI, and SAS, also there are some new developments in SSD Drives which we will come to later, the designation is largely but not entirely derived from the interface used to communicate with the drive.
MFM Hard Disk Drives – These were very early devices, the largest MFM drive was no larger than 100Mb, small by the standards of today, but when these hit the market they revolutionized computing. An MFM Drive can be spotted by the use of two ribbon cables connecting the drive to the PC. They generally had a very particular sound which was the result of steper motors used to move the head across the platter.
IDE (PATA) – Stands for Integrated Drive Electronics, which was a result of a lot of the drive controlling hardware was moved from the controller onto the drive itself. This gave much greater compatibility, MFM drives had to employ specific cards to interface with the computer, these cards knew how to write to specific drives. IDE was a mini revolution in the fact that the controller could now tell the drive “Write this” and the drive itself would work out how to do that. This made drives cheaper, larger, more reliable, and most importantly easy to upgrade. The “P” in PATA stands for parallel which is the way data is sent to the drive. Several data channels on the cable means that information can be sent down each, parallel to the others. IDE Drives had a single ribbon cable which attach them to the PC infrastructure. IDE Drivers were responsible for taking storage space from about 100Mb up to nearly 1Tb.
SATA – Serial ATA, these drives are very much like the slightly older IDE, except that they have a much smaller connector, the information is sent in a serial fashion. SATA drives have taken home computing through the 1Tb barrier. However they still rely on the same principle of a spinning magnetic platter.
SCSCI – SCSI Drives are normally found in servers, they have been available for as long as IDE hard disks and again rely on a spinning medium inside to store data. SCSI Disks generally spin slightly faster, IDE disk usually spin around 5400RPM however some can go up to 7400RPM, SCSI Disks range from 10,000RPM to 15,000RPM. SCSI was the first system to implement RAID or Redundant Array Of Inexpensive Disks, this involved using a number of SCSI Drives to do various Mirroring or Stripping Functions to provide either higher speed access or redundancy in the event of a failure. The most common RAID level is RAID-5 which consists of at least three disk drives, in the event of a failure of a single disk that disk image can be rebuilt using a new disk and the information on the other two drives. This provided a good level of fault tolerance for networked storage space.
SAS – Are the modern serial replacement for SCSI offering smaller footprint drives, higher reliability and cheaper hardware.
SSD – Stands for Solid State Disks, in these the entire substrate and head mechanism has been replaced entirely with electronics so there are no moving parts at all. These give much better power consumption, faster access times, and hardly any weakness with relation to mechanical shock. However they are (correct at the time of writing) very expensive. Usually found in systems which require maximum throughput where cost is not a factor, and very high-end laptops (Sony VGN-TZ32VN) as an example.
MTBF – This stands for Mean Time Before Failure, hard disk drives although technology has made them more resilient, faster, larger, and generally better. However the fact that the manufacturers still quote a Mean Time Before Failure time should be your first clue about the reliability of hard disk drives. They are not, they can fail for many reasons, however when they do the results can be spectacular. Every IT company out there will have a story of someone who did not see the point in backing up, as they had not done it for three years and nothing has gone wrong and then bang, the hard disk fails, and they have lost all of their information. If a hard disk is exposed to mechanical shock the head con hit the platter quite easily the tolerance between the two is microns. When this happens you might not notice any side effects, it could have happened in an area of the disk that may not be is use. If this is the case then the PC may seem normal, until the unit tries to store something in that area of the disk, and then can’t retrieve it. Ok so I lose a file, big deal, well not exactly, see that is maybe ok if it happens to be a user file which is stored there, but now take the example where it is not, maybe it is a registry file. Next time you boot your PC it fails.
What can I do to look after my Hard Disk? Well being extra careful around them is a good start, mechanical shock is the largest killer. The biggest thing to remember is to back up your files no matter how reliable your system is, and how trivial you think this may be, things can change in the blink of an eye.
Platter Clatter – This is the term which has been coined by IT professionals to describe the noise a hard disk makes shortly before the read/write head takes a final, but fatal, nose dive into the platter of the disk. This is a very specific sound which is sometimes a clicking noise or a scraping noise. If the drive can still be accessed this is the time to take a final back up and get the hard disk drive unit replaced. If you continue using the disk drive then chances are at some point within a short space of time it will stop.
There are a lot of data recovery companies out there who can recover data from broken hard disk drives, the Open University suffered a major fire which wiped out a hard disk drive which contained important information, a recovery company was able to recover around 90% of this information with a single shot read process. This is a single shot process which totally destroys the disk being read.
An engineering company in Corby suffered a failure on two RAID drives with in a very short space of time, however a data recovery company was able to recover all of the data from the RAID Array however the process cost a little over £20k.
If your in doubt about your hard disk drive give us a call, we have some tools in our toolbox which will enable us to scan the unit and let you know if there are any problems, most disks can be scanned within an hour, and the cost of replacement hardware should your disk be damaged is very low. So why delay? Check out our website for our contact details: