November 1998--If you've been
following my newsletter, you've probably figured out that I'm a
little obsessive when it comes to disk hygiene and for good reason.
As Steve Gibson says, there are only two kinds of hard drives: Those
that are broken and those that are about to break. Unfortunately,
that's the sad truth. With all the spinning and whirling, your disk
represents the most mechanical, least efficient, and one of the most
fragile parts of your computer.
Before we start on the hard drive, let's think
about how we humans do things to make them more efficient.
Imagine going to the grocery store. How do you
shop? Okay, okay, wrong question: How should you shop? The least
efficient way would be to go to the produce section, pick up some
vegetables, go to the other side of the store and pick up some cat
food, return to the produce section to pick up some fruit and so on.
Instead, you'd (hopefully!) pick up all your items from the produce
section, hit the next section in your route through the store, pick
up the items from that section, and so on. If you're like me, you
probably end up running all over the store anyway, but logically,
you know that's not the most efficient way to do things.
Hard drive data is stored in "sections" in units
called "clusters." The size of each cluster will vary depending on a
number of factors including disk size and format (FAT16, FAT32).
Each cluster holds a chunk of data. Any one file can be stored in
multiple clusters. Your hard drive is full of clusters, all with a
unique "address" that identifies its unique location on the disk
surface.
Here's the deal: The operating system gets a
request to store a file. It looks to the directory, finds the first
available cluster, stores the data in that cluster, has more data to
store, looks for the next cluster, stores the data, and so on.
The problem is that, like a deranged shopping
trip, the operating system acts the same way; it doesn't care where
the cluster is, it just grabs the next available cluster and whack,
it stores the data there.
The end result of this method is that when your
hard drive stores or retrieves data, it can end up jumping around
the disk to read or write each cluster. This condition is called
"fragmentation." Of course your disk drive is fast, but even so, you
can image that having to jump around isn't as efficient as being
able to pick up everything in one contiguous area.
Fortunately Windows comes with a utility that
allows you to "defragment" your hard drive. This utility is called
"Disk Defragmenter" or "Defrag." Defrag runs about and rearranges
files so that all the clusters associated with that file are placed
next to each other. The next time you load that file all the data
clusters are contiguous and the drive doesn't have to jump around
the disk to read the various clusters.
If you have Windows 98, new technology was added
to the defrag program. By monitoring the order in which a program
loads it can arrange files on your hard disk so that they load even
faster. For instance, on my computer, Word 97 loaded almost twice as
fast. This doesn't translate to a lot of real time--disks are pretty
fast--but it can change the way your system "feels" and if it feels
faster, you'll probably be happier.
Running Defrag is easy. Because defrag is doing
disk writes, I recommend that you not run it with other programs
active. Not that this will cause damage, but if defrag recognizes
that data has changed, it will restart and take forever to run. Best
to just wait and run it during a "calm" time.
Run Defrag in either of two ways: Click the start
button, programs, accessories, system tools, and disk defragmenter;
or click start, run and type in DEFRAG. Either method will yield the
same results.
I run Defrag at least once a week, more if I have
the spare time, and would suggest that it become one of your toolbox
essentials.
Back to
article index
