Everything you need to know about a Harddrive

Introduction

This Hub is intended to explain what exactly a harddisk is and what it's specifications mean. I hope this will give people insight into a very important piece of hardware that’s part of a personal computer (PC).

Megabyte (MB) and Megabit (Mbit)



The first thing you need to know about harddisks is how data is calculated. A bit is the smallest unit of information there is. A symbol or signal that can have 2 value’s: yes or no, on or off, high or low, loaded or not loaded. The binary system knows them as 1 or 0. The word “bit” are really 2 English words combined from the words binary and digit.

First the term megabyte. 8 bits together make 1 byte and 1000 bytes together make 1 kilobyte (kB). 1000 KB together make 1 megabyte (MB). And a 1000 MB together make 1 gigabyte (GB). 1000 GB is 1 terabyte (TB). The biggest harddisk available right now is 4 TB (Samsung).

Now there is a other term called megabit (Mbit) which is something different then megabyte (MB). 1000 bits together make 1 kilobit (kbit). 1000 kbit together make 1 megabit (Mbit). And 1000 Mbit make 1 Gigabit (Gbit).

To make a connection between the two I can say 10 kbit make 1 kB. For hardisk data the term megabyte is used. Data communication between a motherboard and a harddisk or internet are presented as megabit.

Sometimes people get confused about what they can get with there internet. Providers make the mistake misusing MB and Mb (Mbit). Let’s say 50 MB is what it says on a internet deal so some people think they get 50 Megabyte which is wrong. You can get 50 Mbit (Mb) which is much less then 50 Megabyte (MB). 50 Mb is calculated like this 1 Mbit = 1024 kbit. So 50 x 1024 : 8 (bits) = 6400 so that means you can almost get 6,5 Megabyte (MB) a second when you download.

What is a harddisk



A harddisk is used to store information (data files) in a magnetic way. And because it’s stored with the use of magnetism the advances are data can be read and written and stored for many years. If some other mechanic device comes close to a harddisk the risk of losing data will become inevitable.

What harddisk sizes are there?



The two most used sizes are 2,5 and 3,5 inch harddisk. 3,5 inch harddisks are use in desktop pc’s and the 2,5 inch are use in laptops/notebooks. Apart from the physical differences in size there is also a difference in capacity. Because the 3,5 inch harddisk has a bigger cylinder more space can fit on it.

Interfaces



The interface is the connection between the harddisk and the motherboard. The faster the connection the more data can be send and received and produced. There are several different connections new and old and they are:

  • PATA (Parallel Advanced Technology Attachment) also called IDE (integrated drive electronics) is an older form of cable and the latest version is called ATA133 (up to 133 Mbyte/s). This technology is obsolete and is being replaced with SATA.

  • SATA (Serial Advanced Technology Attachment) also called serial ATA was designed to replace the older parallel ATA (PATA) standard offering advantages such as reduced cable size, hot swapping, higher transfer rate. The different revisions are SATA revision 1.0 (1.5 Gbit/s), SATA revision 2.0 (3 Gbit/s) and SATA revision 3.0 (6 gbit/s) which also indicate the maximum throughput. In order 150 Mbit/s, 300 Mbit/s and 600 Mbit/s. Revision 3.0 is the fastest and latest edition with the highest throughput to date namely used by Solid State Drives (more on the SSD later).

  • SCSI (Small Computer System Interface) is an older obsolete form of cable usually used in server environments. The latest revision Ultra-640 supports up to 640 MB/s and it’s replaced by SAS. It’s also called parallel SCSI.

  • SAS (Serial attached SCSI) the newest form van SCSI used in server environments. Comes in 2 speeds 3.0 Gbit/s and 6.0 Gbit/s. The SAS connector had many improvements over the parallel SCSI such as SAS supports up to 65,535 devices through the use of expanders, while Parallel SCSI has a limit of 8 or 16 devices on a single channel. SAS has no termination issues and does not require terminator packs like parallel SCSI. Also higher throughput are achieved compared to parallel SCSI and these speeds are on each initiator target connection where parallel SCSI shares the speed across the entire multidrop.

  • USB (Universal Serial Bus) is a way to connect many devices to a pc and has several improved versions. USB 1.0, 2.0 and released in 2008 USB 3.0. USB 3.0 has many improvements foremost the 5 Gbit/s (640 MB/s) of throughput compaired to the USB2.0 (480 Mbit/s). USB 3.0 also reduces the time required for data transmission, reduces power consumption, and is backward compatible with USB 2.0.

  • IEEE 1394 (Institute of Electrical & Electronics Engineers) also know as Firewire is a other serial bus interface standard and external way to connect a harddisk. FireWire S3200 is the latest edition and support 3,2 Gbit/s (400 Mbit/s). Although fireware has less overheat on the cpu and is a true peer to peer network (device to device). USB is more populair although it needs a bus master, typically a PC.

Specifications of a harddisk



RPM of the Cylinders (platters)

Harddisk are comprised of one or several magnetic disks named platters. Those platters contain million of squares where a 1 or 0 can be stored. Those squares are formed by a grid of concentrated circular rings called the tracks and lines running from the center to the outside of the disk. Every square can be manipulated magnetically by the head and they are called sectors. The disks are made of aluminium and have a thin layer of magnetic material on them.

Longitudinal Magnetic Recording (LMR) is an older technique to get magnetic information onto a platter. LMR means that every binaire bit is written horizontally on the platter in there magnetic field. Because the capacity of the disk had to get bigger this LMR technique got into trouble because the bits are right next to each other. The magnetic field of bits next to each other where influencing each other when the density got to high. Write actions could then lead to disortion and corruption of data! Although LMR has been improved over the years it’s limitation is 250 gigabit per square inch on a platter.

PMR (perpendicular magnetic recording) is a new technologie introduced in 2005. The main different with LMR is that the bits in the magnetic field are outlined vertically on the platter. Also with PMR a second less magnetic layer underneath the actual layer is applied where the data is written. The second layer works as a conductor for the magnetic field that is generated by the writing head. It is the guess that 1000 gigabit per square inch can be reached and even more with PMR.

Those platters spin around at a speed that is indicated as RPM (Revolutions per minute). The speed ranges are 4800, 5400, 7200, 10.000 and 15.000. The faster the speed the less latency you’ll have. A normal desktop harddisk runs on 7200 RPM which gives a average latency of 4.16 ms (milliseconds). If you have a harddisk with 15.000 rpm which are usually very high performance desktop harddisks (Western Digital Velocity Raptor) or server disks you get an average latency of 2 ms.

A harddisk capacity can be expanded by added more platters and therefore more read and write heads. In a 3,5 inch housing 5 platters can be placed maximum and in a 2,5 inch housing 3 or 4. So if more data is required to fit on a harddisk the only way to realize this is more data density on each platter. Samsung managed to get 1 TB on a platter so 4 platters x 1 TB make a 4 TB disk.

4k sectors

Sectors have been 512 bytes big for a long time. New harddisks make use of 4 kilobyte sectors. From the year 2011 it has become the standard. It reduces the overhead and more data can be placed on a disk. Every sector has some bytes for address information and at the end some ECC (Error-Correcting Code) information. 4 kb sectors only require 1 ECC block as with 512 bytes those blocks are 8 times more. If you save 4 kb data on a 4k sector disk it will take less physical space. You can place more data on the same space with 4 kb sectors. Most software still uses 512 bytes so harddisk use a emulation to convert to 512 bytes. If the harddisk wants to retain its speed the 8 x 512 emulated byte sectors have to come into place with the first hardware 4k sector. If so the partition is aligned, but if not some very dramatic consequences for the performance of the disk will occure and it’s misaligned. Windows Vista and 7 take account for this where Windows XP does not and the disk is initialized misaligned. There are some tools to correct this in Windows XP and you can download it from here: http://www.wdc.com/global/products/features/?id=7&language=1

Accesstime and Seektime

Disk access time is the time required for a computer to process data from the processor and then retrieve the required data from a storage device. It is determined by a sum of the spin-up time, seek time, rotational latency, and transfer time. Spin-up is the time required to accelerate the disk to operating speed. The seek time measures the time it takes the head to travel to the track of the disk to read or write. Average seek time ranges from 3 ms for high-end server drives, to 15 ms for mobile drives, with the most common mobile drives at about 12 ms and the most common desktop drives typically being around 8-9 ms.

With SSD (Solid State Drive) there are no moving parts, so a measurement of the seek time are electronic circuits preparing a particular location on the memory in the storage device which is much less then a mechanical nature. Typical SSDs will have a seek time between 0.08 and 0.16 ms so you see a very big differents between a conventional harddisk and a SSD. Off course the lower the value of the milliseconds the better/faster it is.

Transfer Rate

The disk transfer rate is the speed at which data is transferred to and from the disk media (actual disk platter) and is a function of the recording frequency. It is generally described in megabytes per second (MB/s). Modern hard disks have an increasing range of disk transfer rates from the inner diameter to the outer diameter of the disk. This is called a zoned recording technique. The key media recording parameters relating to density per platter are Tracks Per Inch (TPI) and Bits Per Inch (BPI). To give an example my Samsung Spinpoint F3 HD103SJ has sequential read/write speed of 130.1MB/s and 162.0 MB/s respectively.

Cache (disk buffer)

Disk buffer also called cache bufferis the embedded memory in a hard drive acting as a buffer, between the rest of the computer and the physical hard disk platter that is used for storage. Modern hard disks come with 8 to 64 MB of memory. The disk buffer is controlled by the microcontroller in the hard disk.

Read-ahead/read-behind

When executing a read from the disk, the disk arm moves the read/write head to (or near) the correct track, and after some settling time the read head begins to pick up bits. Usually, the first sectors to be read are not the ones that have been requested by the operating system. The disk's embedded computer typically saves these unrequested sectors in the disk buffer, in case the operating system requests them later.

Speed maching

The speed of the disk's I/O interface to the computer almost never matches the speed at which the bits are transferred to and from the hard disk. The disk buffer is used so that both the I/O interface and the disk read/write head can operate at full speed.

Command queuing

Newer disks can accept multiple commands while any one command is in operation through command queuing. These commands are stored by the disk's embedded controller until they are completed. Should a read reference the data at the destination of a queued write, the to-be-written data will be returned.

NCQ (Native Command Queuing)

Native Command Queuing (NCQ) is a technology designed to increase performance of SATA harddisks under certain conditions. Internally optimize the order in which received read and write commands are executed. This reduces the amount of unnecessary drive head movement which results in increased performance (and slightly decreased wear of the drive) for workloads where multiple simultaneous read/write requests are outstanding.

S.M.A.R.T. (Self-Monitoring, Analysis and Reporting Technology)

Also written as SMART is a technology to prevent problems and error’s. It detects and reports on various indicators of reliability, in the hope of anticipating failures. When a failure is anticipated the user may choose to replace the drive to avoid unexpected outage and data loss. The manufacturer may be able to use the S.M.A.R.T. data to discover where faults lie and prevent them from recurring in future drive designs. There are various software programs that can read this software like the one I use a lot which is HD Tune.

Solid state drive (SSD)



A solid state drive is unlike a traditional data disk without moving parts. The SSD knows no platters but only memory chips. Because the access time are much faster SSD work faster more power consuming and can take more bumps. The result is a processing speed that is over 250 times that of the standard hard drive, and a far more efficient processor use. In addition to the increase in speed, solid state storage is completely silent because there are no moving parts. With the exception of a cooling fan, SSD create no sound whatsoever. The drawbacks are it’s life span is shorten then of a traditional disk due to memory can’t be written in a unlimited way. Also the price for each gigabyte is higher.

If I take a very populair SSD like the OCZ Vertex 3 120 GB. it reaches max read transfer of 550MB/s and a max write transfer of 500 MB/s. If you compare it to my Samsung harddisks, which are infact very good speeds for a tradional harddisk 130.1MB/s read and 162.0 MB/s write that's a huge difference. If you compare the seek time of my harddisk which is 8.9 ms to the .1 ms of the OCZ SSD that also a huge difference. The .1 is less then 1 ms, now remember the lower the better. Also this ssd has a maximum 4K Random Write of 85,000 IOPS.

IOPS (Input/Output Operations Per Second)

IOPS is the common unit of measurement for I/O performance. The more the better and IOPS can be measured with applications such as Iometer. Iometer is an I/Osubsystem measurement and characterization tool for single and clustered systems. It is used as a benchmark and troubleshooting tool and is easily configured to replicate the behavior of many popular applications. One commonly quoted measurement provided by the tool is IOPS.

What brand of harddisk or SSD to you think is the best?

See results without voting

Speed vs Space

You rather have speed (SSD) or more capacity (HDD)?

See results without voting

More by this Author


Comments 1 comment

Johnny 3 years ago

One thing I might have missed was powering a hard drive from an external enclosure, each one has its own requirement. say my 1tb vs my 3tb one wants (according to the sticker) 5v 420ma 12v 700ma vs 5v .75a 12v .75a what difference does it make? can I change them from enclosure to enclosure? what do I need to learn here?

    Sign in or sign up and post using a HubPages Network account.

    0 of 8192 characters used
    Post Comment

    No HTML is allowed in comments, but URLs will be hyperlinked. Comments are not for promoting your articles or other sites.


    Click to Rate This Article
    working