The last 20 years has brought about a proliferation of computer use in society. Technology is in a constant state of evolution and computer based activities are becoming increasingly integrated with their users’ lives. A number of years ago, it was unthought-of to conduct activities such as banking and shopping online. In addition to this, the vast communication possibilities that are now in existence are making the world a smaller place as people are using a multitude of messaging services such as MSN to stay in contact. The benefits of these highly sophisticated technologies are however, faced by numerous threats. Computers are, more than ever susceptible to malicious use in the forms of hackers, viruses and fraudulent activity. One of the reasons for this is that computers are responsible for storing large amounts of information. Aside from threats posed by malicious use, computers also face a number of issues from an architectural perspective. The vast majority of data stored on a computer is saved on the hard drive and although improvements are continuously being made in terms of their size and capacity, hard drives are always vulnerable to a great deal of stress. In turn, hard drives inevitably have a shelf life of approximately three years. A study conducted by the accounting firm McGladrey and Pullen estimates that one out of every 500 data centres will experience a severe computer disaster this year. As a result, almost of half those companies will go out of business (Peddle 2006).
The objective of this essay is to summarise the main ways in which hard drives can fail, thus incurring a loss of data or difficulty in accessibility. The essay is structured as follows; the first section provides a description of the main ways in which hard drives can fail, along with the effect that this has on the stored data. This is followed by a description of the methods that can be conducted in attempt to try and prevent failure as much as possible and culminating with a brief conclusion and summary of the essay.
Causes of failure A hard disk drive is a digitally encoded non-volatile storage device which stores data on rapidly rotating disks with magnetic surfaces called hard disk platters (Wikipedia 2006). The read-write heads of the hard disk are used to recode and retrieve the data stored on the hard disk as the disk platter rotates at a very high speed. Hard disks operate under extreme stress and will eventually fail due to general wear and tear accrued through age or some form of malfunction or failure. The chances of failure of a hard drive therefore increase greatly over time and ironically the chances are greatly increased as hard disks improve and develop. The phenomenon of hard disk failure is raising higher and higher; as to increase the read and write speed, today we have the latest hard disk rotating amazingly faster and this immense revolving speed generates massive centrifugal force, a single adverse cause in the course of normal operation can cause severe hard disk failure (Bista 2006).
Hard disk failures can be generally classified in two categories, namely physical and logical failures. Physical hard disk failures are the type of failures in which something is physically wrong with the hard drive itself and symptoms may include a grinding or clicking sound (Posey 2006). A logical failure is the type of failure in which there is nothing physically wrong with the drive itself however, the information stored on the hard disk is in bad shape. Normally this type of failure can be repaired by using a tool such as Scan Disk (Posey 2006).
The following section outlines the main cause of hard drive failure which are as follows:
Types of hard disk failure Firmware corruption Firmware is the software code that controls and is embedded in the physical hard drive hardware (Data clinic 2006). It is responsible for the operations of the hard drive and its activities include configuration, management of the interaction that takes place between the other components and the safe powering on/off of the system. At start-up, a hard disk loads the service data from its firmware zone into RAM and then allows the drive to report on readiness. If there is damage to these modules, then the drive will report either it’s family identity or will not detect it at all (Disklabs 2006).
In the event of such an occurrence, the computer will be prevented from accessing any data on the hard disk. The benefit to this is that it is merely an accessibility problem and a firmware failure does not automatically incur a data loss. Once the drive has been repaired, it is possible to recover the data.
Electronic failure Data clinic (2006) describes how electronic failure usually relates to problems on the controller board of the actual hard disk. This may be due to the computer actually suffering a power spike or electrical surge that knocks out the controller board on the hard disk making it undetectable to the BIOS. This means that whilst there is an electronic failure, any data stored on the hard drive will be inaccessible until the problem is fixed. The nature of this type of problem however, is advantageous as it is a mere hindrance and data will not be lost or corrupted and is therefore fully recoverable. Mechanical failure Mechanical failure is generally considered to be more serious than electronic failure or firmware corruption as it often leads to data loss and even complete data loss if prompt action is not taken. There are numerous types of mechanical failure; these can range from motor problems to one of the most common types of known failures is a head crash. A head crash, as it’s name implies, occurs when the read-write heads of the hard disk collide with the disk’s rotating platter surface (Data clinic 2006). Often, this can cause damage to both the heads and even the hard disk itself.
The head normally rides on a thin film of moving air which is entrapped at the surface of the platter (Wikipedia 2006). The platter is coated with a magnetic layer and a crash can occur if a small amount of dust, a sudden movement of the computer or power surge causes the head to bounce against the disk and damaging the magnetic coating. Modern A head crash, as its name implies, occurs when the read-write heads of the hard disk collide with the disks rotating platter surface (Data clinic 2006). Often, this can cause damage to both the heads and even the hard disk itself.
The head normally rides on a thin film of moving air which is entrapped at the surface of the platter (Wikipedia 2006) The platter is coated with a magnetic layer and a crash can occur if a small amount of dust, a sudden movement of the computer or power surge causes the head to bounce against the disk and damaging its magnetic coating. Modern drives spin at extremely high speeds (between 5, 00 and 15,000 rpm, (Wikipedia 2006)) which means that when such a collision occurs, the damage incurred can be quite serious.
Obviously, data corruption will occur at the onset of a head crash as it will then become unreadable when the read-write head attempts to access it. Furthermore, once a head crash has occurred, this increases the probability of future head crashes, especially if there is dust in the system, for instance which means the user needs to take prompt action to rectify this in order to prevent further head crashes and risk losing all data stored on the hard disk.
Circuit Board Malfunction The circuit board is responsible for powering and controlling the mechanical unit. It is exposed to variations in temperature and as a result, the chips that comprise the circuit board surface can develop small breaks internally which can cause circuit interruption and electrical failure. Furthermore, the failure of any of the other electrical components of the computer can result in a defective circuit board, thus causing the hard drive to fail and data access is then prevented.
Logical errors Logical errors also differ from the other failures mentioned as it usually means that there is nothing actually physically wrong with the disk in terms of architectural faults, instead there is a problem with the information stored on it. There could be an invalid entry in a file allocation table for instance (Data clinic 2006). Logical errors are not too serious, unlike mechanical failures and can be tackled with various software tools. Caution should however be extended because if a fix utility is run inappropriately on the hard disk, the stored data can often become unrecoverable. Human error
In a sense it is rare that a hard disk failure could be caused by a human alone. That is, of course providing that users take caution in how they handle a computer. Combined with mechanical failure, human error accounts for almost 75% of all incidents of data loss (Peddle 2006). All of the previously mentioned hard disk failures could be caused by computer misuse on the user’s part. Computers are very sensitive to changes in temperature, dust and dirt etc entering the hard drive, electrical power surges and knocks and bumps which may occur. Users should therefore take extreme caution to prevent any of these circumstances from occurring as much as they possibly can.
Other problems not caused by the hard disk Aside from problems that may occur within the hard drive itself or damage that may be incurred on the users’ part, other failures can occur in other parts of the computer which can have a knock on effect to the performance of the hard disk. These problems include IDE cable malfunction, power connector problems and inappropriate BIOS configuration. Each of these problems is explained below.
It is important that a computers BIOS is configured correctly because if it is not, then it may seem like the hard disk is not working properly. This can be checked by entering the BIOS setup utility and setting the hard disk type to AUTO. This will ensure that the hard disk will be detected automatically when the computer is booted.
Although these problems identified will initially make it seem like there is a problem with the hard drive and the user may feel a great deal of frustration as the cannot gain access to the data they require etc, because there is no actual problem with the hard drive no data will actually be lost. Methods of preventing failure and data corruption The previous section highlighted the main ways in which hard disk failure and data corruption can occur. The following section details some possible preventative measures and courses of action to take with regards to hard disk failure: Back up data
Although this will not prevent a hard disk failure, all information should be backed up to another medium so that in the event of a failure and worse case scenario of a complete data loss, the data can always be recovered from elsewhere. Back up methods can include saving to a DVD, CD, USB memory stick etc, even to another hard disk on another computer. Install two hard disks on the same computer
It is possible to install two hard disks on a computer and place them under a RAID configuration. The term RAID (redundant array of independent disks) means that data can be stored or replicated among the drives so that when one drive fails, the other one works and still holds the data that may have been lost should there have been only one hard drive.
General maintenance As described in the human error section, care should be taken to ensure that computers are not exposed to extreme changes in temperature, they should be kept cool and away from any dust or other contamination that could possibly get inside and cause problems. Care should always be taken when handling and moving computers to avoid any sudden knocks etc. In addition, hard disks have a limited life span, so if the hard drive is only a few years old it would be a good idea to buy a new one.
Sometimes it may seem like the hard disk does not spin. This could be because the heads have become ‘glued’ to the platters; this is quite common in particular in older hard disks. If this happens, turning the computer on and off a few times can often fix the problem. Alternatively the hard disk may appear to stop running all of a sudden. If this is the case it could be a problem with the power connector or it could be more of a serious malfunction. It is therefore a good idea to check all the equipment and it may so be required that a new hard disk is required. Conclusion
This essay has highlighted, first of all the need and importance of computers in today’s society and how there is an increased pressure and stress placed computers from an architectural perspective to store huge amounts of data. The advances in technology mean that the capabilities of computer hard drives are increasing both in terms of speed and storage capacity, but these technological advances cannot unfortunately eradicate the possibility for hard disk failure, which is ultimately inevitable. The benefit is that hard disk failures can be avoided as much as possible though a good knowledge of how they work (i. e. knowing that they will need replacing after a few years of use) and taking care and caution in using them. Furthermore, the majority of hard disk failures do not automatically result in a complete data loss, often the information can be recovered by making the necessary repair or by taking quick action on the fault. Alternatively, it is not possible to completely prevent hard disk failures from occurring and in the case of severe failures such as a head crash data can be rendered completely unrecoverable. Finally, although this can result in what might be deemed as a disaster, backing up data or taking the precautions of configuring the hard drive (e. g. RAID mirror configuration) reduces the impact of the hindrance that such a failure can cause.
B. Posey (2006), ‘Repairing Hard Disk Problems’,
D. Peddle (2006), ‘The human side of data loss: The reality of a hard drive crash’,
Data Clinic (2006), ‘Hard disk failure’,
Data Labs (2006), ‘Firmware data recovery’,
Data Master (2006), ‘Why do hard drives fail?’,
Wikipedia (2006), ‘Disk failure’,
Thanks for reading! - we hope you found this article useful. Chris Seeley, http://www.dataclinic.co.uk
Thanks for reading! - we hope you found this article useful.
Chris Seeley, http://www.dataclinic.co.uk