webpointmorpheus total solution web design

webpointmorpheus Computer Info
Floppy, SCSI, USB, & Hard Drives

Other pages in this series:

©2005 - material compiled by Bob Carnaghi, www.webpointmorpheus.com

Overview     Top of Page
This document treats the means of storage on the computer. Although some of the technology listed here is capable of other applications (SCSI, USB), the technologies are lumped here for convenience.
 
Drive Letter Assignment
When the computer boots, Windows assigns a drive letter or name to hard drives, partitions, floppy drives, and CD-ROM drives. The drive letters A: and B: are reserved for the floppy drives. The first floppy gets the letter A: and the second gets the letter B:, if installed. Hard drives are the next priority, start with the letter C:, and can continue to Z: if necessary. CD-ROM drives by default get the next drive letter after the last installed hard drive. After CD-ROM drives come USB drives, etc.
Floppy     Top of Page
The Floppy technology was at one time the stalwart of the computing industry. Floppy drives are currently being eclipsed by newer, more robust flash drives. In essence, floppy technology has changed very little since the release of the first PC - especially when compared to the advancements of other computer technology. Listed below are some of the basic points of Floppy technology:
  • Floppy disks were originally 5 1/4" wide, and very flexible, which is where they get the name 'floppy'.
  • Floppies use a 34-pin controller cable which often has a 7-wire twist between the connectors that determines the drive letter assignment.
  • If there are two floppy drives connected to the computer, the drive on the end of the controller cable will be drive A:, and the drive in the middle of the cable will be drive B:
  • It is also possible to change the floppy drive letter assignment in CMOS on some computers.
  • Floppy drives are powered by a 'mini' type connector from the power supply.
  • Floppy diskettes are typically 3 1/2" x 3 1/2", double sided, high density, which gives a capacity of 1.44 MB per disk.
  • It is possible to get a double-sided extra-high-density floppy diskette, which doubles the capacity to 2.88 MB per disk.
  • The 34th wire on the floppy cable is called the drive change signal. This wire is responsible for watching the contents of the floppy drive, and changing the display of the contents of the drive if triggered.
  • One will occasionally get the following error message at boot:
    Non system Disk or Disk Error
    Replace and press any key to reboot.
    
    or...
    No operating system found
    
    If this is the case, look first in the Floppy Drive for a non-bootable disk.
  • To create a boot floppy, type the following at the command line: 'format a: /s /q' or 'sys a:'.
  • This method will not work on Windows XP. Format the floppy from Windows Explorer, and choose 'Create MS-DOS Startup Disk'.
SCSI     Top of Page
SCSI (Small Computer System Interface) is a technology for attaching devices to a computer to expand the capabilities of the core hardware on the system. SCSI has been incorporated in computer systems since early in computing history. At the time of this writing, there have been 3 SCSI standards, SCSI-1, SCSI-2, and SCSI-3. In general, SCSI devices can be grouped into two broad categories: internal and external, depending upon their placement relative to the computer tower. SCSI devices can be connected through an onboard SCSI controller or a SCSI card which plugs into an expnasion slot. SCSI devices are NOT FORGIVING with regards to mis-appropriated connections: plugging them in wrong, backwards, or not terminating them properly may destroy the unit as well as the controller.
 
Following is a list of typical SCSI devices, both internal and external:
  1. Hard Drives
  2. Tape backup units
  3. Removable hard drives
  4. Scanners
  5. CD-ROM drives
  6. Printers
 
There are several practical considerations when implementing SCSI devices. Wide SCSI permits up to 16 devices per chain, including the controller. Each device must have a unique ID number within the chain. See the device and/or the device documentation for ID settings. The ends of each SCSI chain must be terminated properly, including the controller if necessary.
 
SCSI-3 has implemented several enhancements to SCSI technology, but has also made understanding the technology very complex. Here are some of the considerations:
  1. Narrow SCSI moves data in an 8-bit path, Wide SCSI moves data in a 16-bit path.
  2. There are several flavors of data transfer speeds: Wide, Ultra, Ultra2, Ultra3, etc.
  3. The number of devices supported depends upon the SCSI standard as well as the architecture of the controller and devices.
  4. The length of each chain depends upon the SCSI standard as well as the architecture of the controller and devices.
  5. There are at least 4 types of termination which depend upon the devices involved.
  6. SCSI devices are capable of Bus Mastering.
 
In the early days of computing, SCSI devices provided a significant improvement over the IDE devices of the day. Small computing systems that wanted/could afford the investment for increased speeds used SCSI. At the time of this writing, developments in EIDE and USB technology has made significant headway and become a cost-effective alternative to the more expensive SCSI devices. However, SCSI devices remain the systems of choice in server and other serious systems.
 
USB     Top of Page
USB (Universal Serial Bus) is a standard of connection architecture which uses a serial bus architecture and provides extremely fast data transfer. At the time of this writing, there are two USB standards: USB 1.1, and USB 2.0. They are also called USB and Hi-Speed USB, respectively.
 
Here are some USB tips and tricks:
  1. Check in CMOS to verify that there is an IRQ used for USB ports.
  2. Windows 95A and Windows NT do not support USB. Windows 95A has a USB supplement available from the Microsoft website.
  3. If the USB device comes with a device driver or setup disk, install it BEFORE plugging in the device.
  4. Windows 2000 and Windows XP provide many USB drivers that are sufficient for many devices, but don't necessarily use all the enhancements of the device. It's best to follow the procedure outlined in #3 above.
  5. USB is Hot-Swappable, which means that the devices can be plugged and unplugged without powering down the system. It is often advisable to use the Eject facility in order to avoid data loss and continuity.

USB Characteristics

Item USB 1.1 USB 2.0
Transfer Rate 12 Mbps 480 Mbps
Hot Swappable Yes Yes
Max devices on a chain
(not always feasible)
127 127
Max cable length
(shorter is often better)
5 Meters 5 Meters
Hard Drives     Top of Page
Hard drives are the backbone of data storage in a computer. By and large, hard drives today come in two flavors, based upon their connection type: PATA (Parallel ATA) and SATA (Serial ATA). (Barring SCSI hard drives, see the SCSI section.) The ATA (Advanced Technology Attachment) is a standard that eclipsed the drives of antiquity (pre-1989) and called for an onboard IDE controller chip that configured the drive to interact with the computer system. These devices are known generically as IDE drives. ESDI (Enhanced Small Device Interface) drives typically require a controller card that contains a system level interface for the drive.
 
Hard drives are measured by two basic criteria: storage capacity and access speed. As technology advances, these criteria grow in leaps and bounds. Five years ago a hard drive of 2 GB was adequate. Today, hard drives can be purchased relatively inexpensively with capacity of 250 GB, no problem.
 
A couple of the most important concepts in hard drive technology are Cylinders, Heads, Sectors per track, clusters, and LBA (Logical Block Addressing). Cylinders reference consecutive rings of space on the circular drive. Heads are the actual devices on the arms that read the data on the drive. A single Sector can contain at most 512 bytes of data, so the sectors are grouped into Sectors per Track (Cylinder) to make data storage more efficient. The original computers were limited to reading data from only a hard drive of 504 MB. LBA permits data to be read from hard drives that exceed this limitation. Low level formatting marks off the disk into cylinders and sectors and defines their placement on the disk. Partitioning is the process of separating the hard drive into distinct managable regions or areas. High level formatting is the process or creating a FAT or MFT for each of these areas.
 
Hard drives (as well as CD-ROMS, etc.) are given their configuration position in the computer device hierarchy by the jumpers on the unit. In addition to the controller to which they are connected on the motherboard (Primary or Secondary) each hard drive can serve as Master or Slave on its respective cable. The position on the cable will determine the unit's order if CS (Cable Select) is jumpered, with the device on the end becoming the Master. The Primary Master position must be a hard drive with a bootable operating system in order to boot from a hard drive.
 
In order to be useful for data storage, a hard drive must be partitioned and formatted for a specific file system. File systems are often OS particular, so the format chosen depends upon the installation of the hard drive unit. When formatted, the hard drive creates a FAT (File Allocation Table) or MFT (Master File Table) that keeps track of contiguous clusters on the drive for data continuity. The most popular Windows file formats are FAT16, FAT32, and NTFS. FAT16 has size and efficiency limitations. Windows NT will not work with FAT32. Of the three, NTFS is the only one that offers any type of data encryption and file security, and is only useful on Windows NT, 2000, and XP.
Hard drives can be partitioned into separate drives for organization. Hard Drive partitions are either Primary or Extended. Extended partitions can be further divided into logical partitions. There is a limit of 4 Primary partitions per drive, and a limit of 24 logical drives per extended partition.
PIO Modes     Top of Page
PIO Modes (Programmable Input/Output) were originally a set of standards that defined a rate of data transfer. PIO modes have been eclipsed by DMA, and are included here for information.

PIO Modes

0 3.3 MB
1 5.2 MB
2 8.3 MB
3 11.1 MB
4 16.6 MB
UPS     Top of Page
UPS (Uninterruptable Power Supply) is the method of ensuring data continuity during power outages. These systems can vary in scope and delivery from small battery type backups to full blown power generators. Listed below are the most common three types which provide backup on a small to moderate scale for a single computer or system of computers.

Types of UPS

Type Description
Online Battery is in the circuit, best for servers
Stand-by (SPS) Battery not in the circuit
Line-Interactive combines best qualities of the other two types
RAID     Top of Page
RAID is a system of Redundant Array of Inexpensive (Indpenendent) Disks. This system is used in environments where large amounts of data are written to disk across a network, or where backup of data is extremely important to eliminate loss. As stated in the acronym, the data is spread across multiple hard drives. The RAID specification calls for different levels of RAID which establish various methods of writing the data and parity (error checking.)

RAID Levels

Level Description Details
RAID0 Striped disk array with no fault tolerance. The several disks operate in parallel for faster data transfer rates.
RAID1 Mirroring and duplexing Creates duplicate copies of data that is split between disks. Mirroring writes from a single controller to two drives. Duplexing uses two separate controllers to write to two different hard drives. Increased access speeds.
RAID2 Data striping w/error recovery Spreads files across several disks; uses large amount of disk space as overhead to maintain parity (error checking.) Not typically used on PCs.
RAID3 Parallel transfer - parity striping Uses less hard drive space for error checking, but data transfer is more time consuming and less efficient.
RAID4 Independent data disks - shared parity disk Interleaves drives to make them appear as one large drive. There is only one parity drive that is shared. Low overhead high maintenance fault tolerance.
RAID5 Independent data disks - distributed parity blocks Popular system where parity is spread across the disks to create accuracy and redundancy.
RAID6 Similar to RAID5 Uses extra disks as well as asynchronous and cached data transmission for guaranteed fault tolerance. Could be called 'Super Raid5'.
RAID10 Very high reliability & performance Uses mirroring and striping. Similar to RAID1 with increased performance.
RAID53 High I/O & data transfer Variation which combines RAID0 with RAID3 to increase performance.
Top of Page
Introduction to Computer Hardware
Boards & BIOS
CPU & RAM
Floppy, SCSI, USB, & Hard Drives
CD & DVD
Sound, Video, & CRT
Intro to Networking
Operating Systems & the Command Line
Windows
Printers
CPU Characteristics
webpointmorpheus Home       Technical Pages Site Map      This page was last modified: Monday November 27, 2006 3:20 PM