Clay Shields, assistant professor of computer science at Georgetown University, explains:
The short answer is: for many reasons. Computers crash because of errors in the operating system (OS) software or the machine’s hardware. Software glitches are probably more common, but those in hardware can be devastating.
The OS does more than allow the user to operate the computer. It provides an interface between applications and the hardware and directs the sharing of system resources among different programs. Any of these tasks can go awry. Perhaps the most common problem occurs when, because of a programming flaw, the OS tries to access an incorrect memory address. In some versions of Microsoft Windows, users might see a general protection fault (GPF) error message; the solution is to restart the program or reboot the computer. Other programming mistakes can drive the OS into an infinite loop, in which it executes the same instructions over and over. The computer appears to lock up and must be reset. Another way things can go amiss: when a programming bug allows information to be written into a memory buffer that is too small to accept it. The information “overflows” out of the buffer and overwrites data in memory, corrupting the OS.
Application programs can also cause difficulties. Newer operating systems (such as Windows NT and Macintosh OS X) have built-in safeguards, but application bugs can affect older ones. Software drivers, which are added to the OS to run devices such as printers, may stir up trouble. That’s why most modern operating systems have a special boot mode that lets users load drivers one at a time, so they can determine which is to blame.
Hardware components must also function correctly for a computer to work. As these components age, their performance degrades. Because the resulting defects are often transient, they are hard to diagnose. For example, a computer’s power supply normally converts alternating current to direct current. If it starts to fail and generates a noisy signal, the computer can crash.
The random-access memory (RAM) can err intermittently, particularly if it gets overheated, and that can corrupt the values the RAM stores at unpredictable times and cause crashes. Excessive heat can crash the central processing unit (CPU). Fans, which blow cooling air into the computer’s case, may fail, making components susceptible to overheating. And they push dirt and dust inside, which can lead to intermittent short circuits; compressed air or a vacuum cleaner easily gets rid of such dirt. Still other hardware problems, including a failed video or network card, are trickier to identify, requiring software tests or the sequential replacement of components.
Errors on a computer’s hard drive are the most intractable. Hard disks store information in units called sectors. If sectors go bad, the data stored on them go, too. If these sectors hold system information, the computer can seize up. Bad sectors also can result from an earlier crash. The system information becomes
corrupted, making the computer unstable; ultimately the OS must be reinstalled. Last and worst, a computer can fail completely and permanently if the machine gets jarred and the head that reads information makes contact with the disk surface.
Saturday, March 13, 2010
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment