![]() |
![]() |
![]() |
| ![]() |
|
Hard DrivesYou can have the fastest video card out there, 256 megs of SDRAM, 1 meg of Burst cache, 21 inch monitor, 450 Pentium 2 processor, the works; but it matters little if you skimped on the Hard Drive. The Hard Drive is the Achilles heel of any computer because it's the slowest peripheral you have. In this article we'll go over Hard Drives, differences in EIDE and SCSI, terms, File systems, and, of course, things that should be known specific to OS/2 users. The BasicsOn a hard drive, data is stored in the magnetic coating of the platters. The so called head, held by an actuator arm, is used to write and read data (like a record player). This disk rotates with a constant turn time, measured in revolutions per minute (rpm). Data is organized on a disk in cylinders, tracks and sectors. Cylinders are concentric tracks on the surface of the disk. A track is divided into sectors. And a hard drive has a head on each side of a disk. Nowadays, the actuator arm is moved by a servo-motor (not a step-motor which needs more time while swinging in after moving over the desired track). All hard drives have reserved sectors, which are used automatically by the drive logic if there is a defect in the media. Typical hard drives have a rotation speed from 4,500 to 7,200 rpm, with 10,000 rpm drives just now getting to the market. The faster the rotation, the higher the transfer rate, but also the louder and hotter the drive. You may find you need to cool a 7200 rpm disk with an extra fan to prevent shortening the drive's lifespan. Modern drives read all sectors of a track in one turn (Interleave 1:1). The rotation speed is constant. All modern hard drives also have their own cache varying in size and organization. The cache is normally used for writing and reading. On SCSI HD's you may have to enable write caching, because often it is disabled by default. This varies from drive to drive. You may be surprised that it is not the cache size that is important, but the organization of the cache itself (such as a read/write cache or a look-ahead cache). With most EIDE drives, the PC's system memory is also used for storing the HD's firmware (e.g. software or "BIOS"). When the drive powers up, it reads the firmware from special sectors. By doing this, manufacturers save money by eliminating the need for ROM chips, but also give you the ability to easily update your drives "BIOS" if it is necessary. Transfer rate is higher when data is read or written to the outer parts of a disk. The simple reason is that there is more space there for sectors to be read in one turn. The number of sectors varies in steps. Usually on a disk there are 10 to 20 zones (called 'notches') with a constant sector number. That's the reason why you see the steps in the transfer rates. Some hard drives use the combination of 'vertical' and 'horizontal' mapping. The 'horizontal' mapping is used in the zones, the 'vertical' mapping between the zones. However, transfer rate and seek time look the same to 'vertical' mapping. The Interface (EIDE / SCSI)Currently there are 2 different interfaces: EIDE and SCSI. You will often find an EIDE controller already integrated with the motherboard and that EIDE drives are much cheaper than SCSI drives. For SCSI you'll usually need an extra controller, because there aren't a lot of motherboards with them integrated. Together with the higher price of a SCSI disk, a SCSI system is more expensive than EIDE. With EIDE you have a primary and a secondary channel that will connect to two devices each for a total of four per machine. Those could include a mix of devices such as a hard drive, CD-ROM or disk changers. Lately there have also been tape backups with EIDE connectors, but you need special backup software to use them. OS/2 benefits the most from SCSI, since it'll make simultaneous accesses more frequently - a task that SCSI handles better than EIDE. If you have a server or are working with large files like audio, video or other disk-intensive applications, you will benefit more with SCSI. There are three reasons for this:
If you need large capacities and the highest transfer rates available on the market you need SCSI. This is not because EIDE is incapable of this, it's because of the market. High-end disks with high capacities and high performance are intended to be used in servers and aren't built with EIDE interfaces. At the moment, EIDE disks are built with up to a 25 Gigabyte capacity (there is a problem with a 4 GB barrier in some BIOSes, and for drives bigger than 8 GB you need a new BIOS that supports the INT 13 functions AH=41h bis 49h. This is why there were refreshes in the Install Diskettes for OS/2), and transfer rates of about 9 MBytes per second. IBM just came out with new drivers for drives over 8.4 gigabytes. I personally have a 9.1 gig drive and are not having any problems. If you need more, you'll have to use SCSI. Also, SCSI drives have larger cache RAM than EIDE drives do. New news is that IBM is coming out with a new Hard Drive (EIDE), that will be 25 Gigs. If that doesn't get you, then wait for Seagate's 50 Gig drive! The need for speedYou need to know how a slow or fast hard drive affects your overall system performance in a standard environment. If your operating system isn't constantly swapping (e.g. you have enough memory) the speed of a hard drive is only a small part of a well balanced system. Let's say you have a drive that has 30% better performance than another older one; the benefit for standard applications would be from 2% up to 18%. Sometimes, you want or need the fastest components available. Other times, more capacity and reliability is needed. There are several programs available that test the performance of a hard drive. Some are so-so, others are good. In any case, if you have one, you get numbers that tell you something. But do you have a point of comparison? Different benchmarks mean different numbers. Different environments mean different numbers. Modern benchmarks are independent from existing data on the drive (but only read performance testing can be done). But a benchmark could be affected by several things:
A good program to try is SysBench 0.9.4b. If you run it, send me the results and I will compile them and have the results in my next article. Please run all of the tests if you do. If you're interested in my own hard drive results, here they are (.GIF, 4K). File systemsIf you purchase a new hard drive you'll find that it is physically pre-formatted. This means that the cylinder, track and sector information is already written onto the disk. You now have to partition the disk to prepare it for the logical formatting of the file system and writing the partition information and boot sector to the disk. You don't have to use the whole disk with one partition. You can divide it into several, and depending on the operating system there are several file systems to choose from. The most used file systems are FAT (DOS, Windows, OS/2), NTFS (Windows NT) and HPFS (OS/2). I will only be talking about FAT and HPFS, as these are the ones that come built into OS/2. FAT: File Allocation TableDOS, including DOS 7.0 of Windows95/98, and OS/2 can use the FAT file system to store data to floppies and hard drives. The FAT organizes multiple sectors into clusters and uses 12 or 16-Bit cluster address numbers. The 16-Bit FAT is able to address up to 65526 clusters (some are used for special purposes). A cluster can be as big as 32K, which translates into a maximum partition size of 2 Gigabytes. The side effect is that every file takes up at least 32K worth of disk space, even if it is only one K in size. That is the reason why power users forced to use FAT divide a hard disk into slightly less then 512 Megabytes per partition -- then only 8 K per cluster are used and you don't waste too much of your limited space on. If you want to know the size of the clusters on your FAT formatted partitions, you can check by using chkdsk.exe at the OS/2-prompt. If you have a hard drive that is larger then 2 gigs or you use OS/2 exclusively then don't use FAT! It was designed two decades ago for floppy disks, modified later for 10 megabyte hard drives, and looses on average 15% of your disk space to "slack" or leftover cluster space. Instead, you need to use... HPFS: High Performance File SystemA standard OS/2 Warp 4 installation supports two file systems for use on your hard disks: FAT and HPFS. The HPFS file system supports file and directory structures different from the FAT file system and is much better than FAT as a result, although FAT is still retained for backwards compatibility and diskettes. HPFS gets better performance from the intelligent positioning of data on the drive (starting from the physical center of the disk and spreading outwards, so the drive head has less distance to move), plus it allows real long file-names, is less likely to lose data, more resistant to file fragmentation, and thanks to storing data by the sector instead of in clusters it uses disk-space more efficiently than FAT. Another advantage of HPFS over FAT is that it's a "forking" filesystem, storing extra data about files and directories in an area called the Extended Attributes (EAs). For example, the name and icon of an object that appears in an OS/2 folder or on the OS/2 Desktop is stored in its EAs. In HPFS, EAs are part of the HPFS file control lock which is read when the file is opened. In FAT, EAs are stored in a separate file in separate clusters and require additional I/O to access them, and are therefore slower. HPFS has two limitations: Native DOS applications can't see HPFS formatted disks (although DOS programs running in OS/2 Warp can and there exists a driver to read HPFS disks from plain DOS), and the HPFS driver takes approximately 264KB of memory. For more info on HPFS you can jump to IBM's page. One note is that Warp 5, or Warp Server for e-business, will have the Journaling File System (JFS). While I have a Beta of Warp 5, I'm having problems intalling it. What I will say about JFS is the same things that IBM themselves talk about. Some of major subjects JFS tackles are:
You can read more on this at IBM's site. Hard Drives: Who's out there?The big makers of hard drives are IBM, Seagate, Western Digital and Quatum, with some input from Fugitsu and Maxstor. With so many decisions on what hard drive to buy, look first at how much you want and how much you want to spend. If your running OS/2 and another OS you might want at least 6 gigs of drive space. Why? Remember the Hard drive is one of the most painful things to upgrade because all your important data is stored there. The last thing you want is to run out of drive space 6 months down the road. With the extra space you don't have to worry about being "space deficient". My first hard drive was a whopping 250 megs. My brother laughed at me because he thought I would never fill it up. In four months I filled it up and used Stacker 4.0 to double it, two months latter I filled that space also. Speed is important, like before, the slowest part of any computer is the hard drive. Remember: The faster the drive the faster the computer as a whole. Next time we will be going over RAM which goes hand-in-hand with Hard drives. |
![]() |
![]() |
![]() |
![]() |
Copyright © 1998 - Falcon Networking | ISSN 1203-5696 | December 16, 1998 |