|
|
F.A.T. {file allocation tables}
EXAMINING THE FAT 32 SYSTEM
If you're like most Windows 95 users, you're running a Pentium system with a gigabyte or more of hard disk space designated as drive C. Since one G13 is equivalent to 1,024MB, you probably never imagined that you'd fill your disk to capacity. However, you no doubt noticed that your hard disk appeared to fill up much quicker than you thought possible.
* A large percentage of your hard disk is probably empty.
This space is wasted and unavailable because the current FAT (file allocation table) system in Windows 95 wasn’t designed for today's massive hard disks.
FAT doesn't manage the available space on large hard disks very efficiently.
To make Windows 95 use today's hard disks more efficiently, microsoft upgraded the FAT system.
* Known as FAT 32
* Is now available with the OSR2 version of Windows 95(B)
* Standard equipment with Windows 98 when it ships.
WHAT IS THE FAT AND HOW DOES IT WORK?
Imagine that your hard disk is a filing cabinet. In a filing cabinet, you store information on pieces of paper as files. You store the files in folders within folder dividers. To keep track of what's in the folders and dividers, you place tabs on them noting their contents.
Basically, the FAT system works the same way with your hard disk. It provides the electronic equivalent of papers, folders, dividers, and tabs.
The FAT system stores the tabs indicating the file locations in a special database called the
file allocation table, or FAT for short.
* On a hard disk the pieces of paper are known as sectors.
Sectors are the smallest units of storage on your hard disk and measure 512 bytes in size.
To keep track of the sectors, the FAT organizes them into units called clusters, which are similar in function to the file folders.
* When you save a file on your hard disk, the operating system looks in the FAT for free space in the cabinet.
* Writes the data to the sectors
* Makes notations of the filename and the cluster's location in the FAT.
*The problem with the FAT is that its limited in size.
*It can store only so many items at a time.
Some History of FAT
* The FAT system was invented in 1977 as a way to store data on floppy disks for Microsoft Standalone Disk Basic.
* At that time, floppies stored 180KB of data.
* When FAT was invented, it was capable of holding only 64KB of data.
* This 64KB limit was fine for the storage capacities at the time.
* FAT had reached its limit.
* With 65,535 sectors of 512 bytes, the maximum hard disk size that the FAT could keep track of was 32NM (65,535 * 512).
* This meant that if you had a large disk, you had to partition it into 32MB pieces in order to stay within FAT's limits.
For example, to use a 40MB hard disk, you'd have to partition it into two drives: one would be 32MB and the other would be 8MB.
* To compensate for the 64KB limit on the FAT, Microsoft developed a workaround which it implemented in MS-DOS 4.0.
*Extended the FAT's capabilities
* It grouped the sectors on the hard disk into clusters and then replaced the sector addresses in the FAT with cluster addresses.
* FAT could store addresses for 65,535 clusters.
* Could now create partitions larger than 32MB and still stay within FAT's 64KB limit.
* When you create a partition in DOS that's larger than 32MB, DOS increases the number of sectors contained in a cluster in order to keep the FAT under its 64KB limit.
Table A shows the sizes of the clusters used by different hard-disk sizes.
Table A: The bigger your disks, the bigger your clusters
Disk Size Cluster Size
0 MB - 32MB 1/2 KB
33MB - 64MB 1KB
65MB - 128MB 2KB
129MB - 256MB 4KB
257MB - 512MB 8KB
513MB - 1GB 16KB
1.1GB - 2GB 32KB
2.1GB - 4GB 64KB
FAT = fat clusters
Table A shows as hard disk got larger, DOS was forced to increase the cluster size to remain within the 64KB limit.
This system worked fine until hard disks surpassed the gigabyte level increasing the cluster sizes to 32KB.
Having such a large cluster size can be a huge problem for most users--especially if you store lots of small files on your hard disk.
* No matter how small the file, it will consume at least one cluster.
*For example, if you store a 1 KB file on a 1.2GB hard disk, that small file will take up an entire 32KB of space.
* The other 31KB will remain empty.
* This unused space in a cluster is called slack space.
* Since the operating system can't write data into slack space, it's completely wasted.
How is FAT 32 MORE EFFICIENT? FAT 32 is still based on the original FAT system and works very similarly in order to remain compatible with existing programs, networks, and device drivers.
The biggest improvement in FAT 32 is its ability to efficiently manage storage space on today's large hard disks.
SMALLER CLUSTER SIZE
* To improve storage efficiency, the FAT 32 system uses a 4KB cluster size for all hard disks under 8GB.
* This reduces the amount of slack space on your hard disk when you save small files.
* Saving a 1KB file on a IGB hard disk using the old FAT system takes up 32KB of space.
* Saving the same file on the same hard disk using the FAT 32 system, however, takes up only a 4KB of space—that’s a savings of 28KB.
* Guarantees that the average large hard disk will use its disk space at least 10 to 15 percent more efficiently.
* Space savings is even larger--reaching almost the 50-percent range.
* Results vary depending on the number of small files on the hard disk.
Improved reliability
FAT 32 also offers advantages that fall into the category of improved reliability.
* Under the FAT 16 system, the root directory could be located only at the beginning of the hard disk.
* If anything happened to that section of the hard disk, the whole thing was unusable.
* Under the FAT 32 system, the root directory can be located anywhere on the hard disk.
* If something does happen to the section of the hard disk storing the root directory, the new FAT 32 hard disk utilities will be able to easily move the root directory and repair the defective area.
FAT 32 can use both the default and the back-up copy of the FAT.
The FAT 16 system can use only the default copy to run your system; the backup copy is used only by low-level disk utilities when repairing the default FAT.
* FAT 32, on the other hand, can use both copies of the FAT.
* If something happens to the default FAT, the system will continue to run off the backup copy until the default can be repaired.
A LARGER ROOT DIRECTORY
The FAT 32 system has eliminated the 512 entry limit of the root directory.
* The FAT 16 system uses 32 sectors of 512 bytes each to store the root directory.
* The size of the root directory on a hard disk is limited to 16KB, or 512 entries.
* The FAT 16 system will let you store at most 5 12 files and folders in the root directory.
* Under the FAT 32 system, you can have as many files and folders in the root directory as you want.
LARGER HARD DISKS
Microsoft has given FAT 32 the ability to support hard disks as large as 2TB.
* A terabyte is equal to 1,099,511,627,776 bytes--roughly one trillion bytes!
MAXIMIZING HARD DISK STORAGE CAPACITY
* If you're not running the OSR2 version of Windows 95 (version 4.00.950b) or Windows 98
* Other ways of reducing the amount of slack space on your large hard disk.
* Basically, you need to reduce the size of the clusters.
* To accomplish this, you have two options:
1 . You can compress your hard disk by using Windows 95's DriveSpace utility.
2. You can partition your hard disk into two or more drives.
Disk compression is probably the easiest because you don't have to reformat your hard disk or reinstall your applications.
* Remember that the amount of slack space on your hard disk is determined by the types of files you store there.
* If the majority of the files you store on your hard disk are large, such as databases or CAD drawings, you probably don't have much slack space on your hard disk.
* Most computer users store a lot of small files, such as word processing documents or spreadsheets having more slack space.
DETERMINING SLACK SPACE
* Chkdsk, can be use to estimate the amount of slack space your hard disk has.
* Running Chkdsk provides you with useful information about your hard disk, such as total
disk space, the number of files, free space, and disk capacity.
* The Chkdsk utility also tells you the size of your hard disk's clusters.
* After you run Chkdsk you'll find information about the clusters, which Chkdsk calls
allocation units.
* The cluster size number represents the size of the clusters in FAT.
* The second line tells us the total number of allocation units on the hard disk, or the
maximum number of entries that the FAT can hold for our particular hard disk.
* You can estimate the amount of wasted space on your hard disk by doing a little math.
* To estimate the amount of disk space in use by your files, take the information that
Chkdsk provides and plug it into the following formula:
(of hidden files + # of directories + # of user files) * (# of bytes in an allocation unit / 2)
This formula calculates the slack space by adding the amount of entries on the hard disk
and multiplying that total by half of the allocation unit size.
* If you want an exact measurement of the amount of slack space on your hard disk, you should use Norton Utilities for Windows 95.
The Norton System Doctor utility has a slack space sensor that monitors the slack space on your hard disk and displays the results in its window.
USB: THE NEXT PERIPHERAL COMMUNICATIONS CHANNEL
As you know, advancement in computer hardware is a never ending process.
There's a new technology on the horizon that's poised to dramatically increase the speed with which peripheral devices communicate with the PC.
This new technology is called the Universal Serial Bus (USB) communications channel.
Many of the new systems arriving on the market today feature USB ports.
* In addition, Microsofts newest Windows operating system, Windows 98, will provide direct support for the USB communications channel. (windows 95{B} already supports USB)
In the coming months, you'll also begin to see USB devices on the market.
Once this hardware and the new operating system take hold, the possibilities are limitless.
The USB will replace the parallel and serial communications channels.
THE PARALLEL CHANNEL
* Most of us rely on the parallel channel for fast access to printers and other devices, such as external tape and CD-ROM drives.
* A standard parallel cable consists of 25 wires that transfer information one byte at a time.
* Each wire terminates in a connector called a pin, so we often refer to a standard parallel cable as a 25-pin connector.
* The 8 bits that make up each byte travel in parallel across 8 of the 25 wires.
* The parallel channel is fast and reliable and has long been the preferred way to connect printers and other external devices to a computer.
* One limitation of the standard parallel channel is the distance you should place between the device and the computer.
* The parallel channel is best suited for
distances of 15 feet or less.
Longer parallel cables are available, but the extra length may
corrupt the data unless the signal is amplified.
The vehicle of this channel, the parallel port, has undergone many transformations since the first IBM PC hit the market.
Port Type Used With
Unidirectional 4-bitand compatibles Original IBM PCs
Bidirectional 8-bit and compatibles EBM PS/2s
Type 3 direct memory access (DMA) Later PS/2s and compatibles
Enhanced parallel port (EPP) computers 386-based
Enhanced capabilities port (ECP) Pentium-based computers
* The original IBM PC incorporated a unidirectional parallel port, which as the name implies, can send information in only one direction.
* Bi-directional ports simply used an additional 8 of the 25 wires in the parallel cable for data traveling in the other direction.
For instance, bi-directional ports enabled your printers to send status messages about print jobs back to the computer.
* Type 3 direct memory access (DMA) ports--allowed much faster bi-directional performance.
* This technology relied on the computer setting aside a block of memory to hold or cache data heading for the parallel port.
* Type 3 DMA ports improved performance dramatically, since the CPU no longer needed to regulate the flow of information.
* The next generation of parallel port technology, the enhanced parallel port specification (commonly referred to as EPP, or fast mode port), depends on intelligent peripheral devices that can dynamically manage the information traveling across the cable.
* The current EPP standard is also commonly referred to as IEEE 1284.
* The newest addition to the world of parallel ports is the enhanced capabilities port (ECP).
Parallel ports using this technology surpass the performance of EPP ports, but to do so, they require software designed to take advantage of ECP's direct memory access and data compression capabilities.
THE SERIAL CHANNEL
* The serial channel is the primary means for connecting modems and mice to the serial port in your system.
* The word serial refers to the fact that the data is sent in series, one bit at a time, over a single wire.
This design is significantly slower than sending 8 bits at a time via a parallel channel, but serial signals can travel much farther without degradation.
* Serial ports are only as good as their universal asynchronous receiver-transmitter (UART) chips.
* These chips are the core of a serial port's operation.
* Basically, this chip's job is to convert data heading out through the serial port from the computer' s native parallel format into a serial stream of information.
* The UART is also responsible for reassembling data coming in through the serial port and converting it back into the parallel format expected by your computer's data bus.
* The various UARTs you're likely to encounter.
Table B: UART evolution
UART Systems
8250 PC, XT, and AT
164501 80386
16550 180486 and Pentium
* The older PC- and XT-class systems and some inexpensive serial cards use the 8250 UART, which is reliable, but it doesn't support connections of 9,600 or more bits per second (bps).
* The 8250 chip has a number of known bugs, but
many computers were designed to anticipate these flaws and can
compensate for them.
A number of 80386-based computers use the 16450 UART. Because
this chip had a larger data buffer than its predecessor, it both
improved overall performance and fixed several bugs.
The next UART chip to hit the market was the 16550.
One of the major new features of the 16550 is that it has a 16-byte buffer and uses the first in, first out (FFO) buffering technique.
The 16550 can continue to receive incoming characters and store them in the buffer while the CPU is busy handling other tasks.
Since it can continue to work while the CPU is busy, the 16550 has brought major performance gains in serial communications.
Over time, the 16550 was improved upon and reincarnated as the 16550A, 16550AN, and then the 16550AFN.
THE USB CHANNEL
The future of PC communications channels-the Universal Serial Bus.
A USB port will functionally replace all the ports now found on the back of your computer.
Using a single cable, a USB port will allow you to connect keyboards, mice, joysticks, scanners, printers, monitors, telephones, modems, ISDN modems, and an assortment of other devices to your computer.
You can daisy-chain up to 127 devices from a single port, since each length of USB cable can be as long as 5 meters (more than 15 feet).
The USB channel pumps a lot more information through the cable than just data; it also supplies a 5-volt power line to the peripherals connected to it.
Not only will USB do away with the many cables used to connect peripheral devices to the system but it will also eliminate the multitude of power cables dangling from the back of your desk to the power supply on the floor.
USB ports are incredibly fast.
USB has two data speeds: 1.5Mb per second (Mbps) for lower-end devices (such as keyboards, mice, and joysticks) and 12Mbps for higherend devices (such as scanners, printers, monitors, and modems).
To put this into perspective, 12Mbps is comparable to a lOBaseT Ethernet network's speed.
Best of all, when using an operating system that supports it, USB works very much like the current Plug-and-Play system now found in Windows 95, allowing you to add and remove devices without powering down or reconfiguring the computer.
When a new device is added or removed, the system automatically detects the change and then loads or unloads the appropriate driver.
PERFORMANCE FACTORS AND TRADEOFFS IN CONFIGURING FOR MULTIPLE DEVICES
Master/Slave Channel Sharing: By its very nature, each IDE channel can only deal with one request, to one device, at a time. You cannot even begin a second request, even to a different drive, until the first request is completed. This means that if you put two devices on the same channel, they must share it. In practical terms, this means that any time one disk is in use, the other must remain silent. In contrast, two disks on two different IDE channels can process requests simultaneously on most motherboards. In general, you get better performance putting each drive on its own channel. (This restriction is one major disadvantage of IDE compared to SCSI.
Independent Master/Slave Device Timing: Some chipsets allow two devices on the same channel to use different PIO modes, but some do not. If you are using two devices with radically different maximum transfer rates, and the chipset doesn't support independent timing, you will slow down the faster device to the speed of the slower one.
Hard Disk and CD-ROM Channel Sharing: There are several reasons why CD-ROMs (or other ATAPI devices) should not be shared on the same channel as a fast hard disk. ATAPI allows the use of the same physical channels as IDE/ATA, but it is not the same protocol; ATAPI uses a much more complicated command structure. CD-ROMs are also generally much slower devices than hard disks, so they can slow a hard disk down when sharing a channel. Finally, many CD-ROMs cannot deal with DMA bus mastering drivers, and will cause a problem if you try to enable bus mastering for a hard disk on a channel they are using.
Older Hard Disks: Older hard disks, typically those three years in age or more, typically only support the older, low speed PIO modes for transfers (check the drive's specifications to be sure). In this case, the caveats about hard disk performance matter much less, since the drive is relatively slow anyway.
Interface Bus Type: High-speed transfer modes require the use of a local bus (PCI or VLB)
hard disk controller. ISA-based controllers, such as regular sound cards that are commonly used as tertiary IDE channels, are not suitable for use with high-speed modem disks.
IRQ Resource Usage: Each IDE channel requires an IRQ line, and in some PCs these are a
scarce commodity. In some cases devices can be combined on a single channel with
minimal impact on performance, allowing the reclamation of one or more IRQs for
use by other peripherals.
REDNECK COMPUTER TERMS
BACKUP - What you do when you run across a skunk in the woods
BAR CODE - Them's the fight'n rules down at the local tavern
BUG - The reason you give for calling in sick
BYTE - What your pit bull dun to cusin Jethro
CACHE - Needed when you run out of food stamps
CHIP - Pasture muffins that you try not to step in
TERMINAL - Time to call the undertaker
CRASH - When you go to Junior's party uninvited
DIGITAL - The art of counting on your fingers
DISKETTE - Female Disco dancer
FAX - What you lie about
IRS HACKER - Uncle Leroy after 32 years of smoking
HARDCOPY - Picture looked at when selecting tattoos
INTERNET - Where cafeteria workers put their hair
KEYBOARD - Where you hang the keys to the John Deere
MAC - Big Bubba's favorite fast food
MEGAHERTZ - How your head feels after 17 beers
MODEM - What ya did when the grass and weeds got too tall
MOUSE PAD - Where Mickey and Minnie live
NETWORK – Scoop’n up a big fish before it breaks the line
ONLM - Where to stay when taking the sobriety test
ROM - Where the pope lives
SCREEN - Helps keep the skeeters off the porch
SERIAL PORT - A red wine you drink with breakfast
SUPERCONDUCTOR - Amtrak's Employee of the year
SCSI - What you call your week-old underwear