Hard disk drive ( HDD ), hard disk , hard drive or fixed disk is an electromechanical data storage device that uses magnetic storage to store and retrieve digital information using one or more fast rotating disks (disks) coated with magnetic material. The plates are paired with magnetic heads, usually arranged on a moving actuator arm, which reads and writes data to the surface of the platter. Data is accessed by random access, which means that each data block can be stored or retrieved in any order and not just in sequence. HDD is a non-volatile storage type, retaining the stored data even when it is turned off.
Introduced by IBM in 1956, HDD became the dominant secondary storage device for general purpose computers in the early 1960s. Continuously enhanced, HDD has maintained this position into the modern era of servers and personal computers. More than 200 companies have been producing HDD historically, although after extensive industry consolidation most of the units are manufactured by Seagate, Toshiba, and Western Digital. HDD dominates the volume of storage generated (exabytes per year) for the server. Although production is growing slowly, sales revenue and unit shipments decrease as solid-state drives (SSDs) have higher data transfer rates, higher storage area density, better reliability, and much lower latency and access times.
Revenue for SSDs, which mostly use NAND, is slightly more than HDD. Although SSDs cost almost 10 times higher per bit, they replace HDD where speed, power consumption, small size, and durability are important.
The main characteristic of HDD is its capacity and performance. Capacity is determined in the unit prefix corresponding to the power of 1000: 1-terabyte (TB) drive has a capacity of 1,000 gigabytes (GB: where 1 gigabyte = 1 billion bytes). Typically, some HDD capacity is not available to users because it is used by file systems and computer operating systems, and possibly built-in redundancy for error correction and recovery. Performance is determined by the time it takes to move the head to a track or cylinder (average access time) plus the time required for the desired sector to move below the head (average latency, which is a function of physical rotation speed in revolutions per minute ), and finally the speed of data transmission (data rate).
The two most common form factors for modern HDDs are 3.5 inches, for desktop computers, and 2.5 inches, especially for laptops. HDDs connect to systems with standard interface cables such as PATA (Parallel ATA), SATA (Serial ATA), USB cable or SAS (Serial Attached SCSI).
Video Hard disk drive
Histori
The first IBM hard drive production, 350 disk storage shipped in 1957 as a component of the IBM 305 RAMAC system. That's roughly the size of two medium-sized refrigerators and stored five million six-bit characters (3.75 megabytes) on a 50-disk stack.
In 1962, the IBM 350 was replaced by an IBM 1301 disk storage unit, consisting of 50 platter, each approximately 1/8-inch and 24-inch diameter. While the IBM 350 uses only two read/write heads, 1301 uses an array of heads, one per plate, moving as a whole. The read/write operation of the cylinder mode is supported, and the head flies about 250 micro-inches (about 6 Âμm) above the platter surface. The motion of the array head depends on the binary adder's hydraulic actuator system which ensures repeating positions. The 1301-sized cabinet about three home refrigerators are placed side by side, saving the equivalent of about 21 million bits of eight bytes. Access time is about a quarter of a second.
Also in 1962, IBM introduced a 1311 model disk drive, which is about the size of a washing machine and stores two million characters on a removable disk package. Users can purchase additional packages and change them as needed, such as magnetic tape reels. Then removable drive pack models, from IBM and others, became the norm in most computer installations and reached a capacity of 300 megabytes in the early 1980s. Non-removable HDD is called a "fixed disk drive" drive.
Some high-performance HDDs are made with one head per track (for example, IBM 2305 in 1970) so that no time lost physically moves the head to a track. Known as fixed-head or head-per-track disk drives they are very expensive and no longer produced.
In 1973, IBM introduced a new type of HDD code called "Winchester". Its main differentiator feature is that the disk head is not completely pulled from the disk disk stack when the drive is turned off. Instead, the head is allowed to "land" in a special area of ​​the disk surface when the spin-down, "take off" again when the disc is then turned on. This greatly reduces the cost of the head actuator mechanism, but prevents the removal of only disks from the drive as it does with the disk packs of the day. Instead, the first "Winchester Technology" drive model displays a removable disk module, which includes the disk pack and head assembly, leaving the actuator motor in the drive after it is released. Then the "Winchester" drive abandoned the removable media concept and returned it to a non-removable plate.
Like the first removable drive pack, the first "Winchester" drive uses a 14-inch (360 mm) diameter platter. A few years later, the designers explored the possibility that physically smaller platters might offer an advantage. Drive with an inalienable eight-inch dish appears, and then a drive using 5 1 / 4 Ã, in the form factor (130 mm) (mounting width equivalent to that used by contemporary floppy disk drives). The latter is primarily aimed at the personal computer market (PC) which then develops.
As the 1980s started, HDDs were a rare and very expensive additional feature on PCs, but by the late 1980s their cost had been reduced to the point where they were standard on all but the cheapest computers.
Most HDDs in the early 1980s were sold to PC end users as additional external subsystems. The subsystem is not sold under the name of the drive manufacturer but under the name of a subsystem manufacturer such as Corvus Systems and Tallgrass Technologies, or under the name of a PC system manufacturer such as Apple ProFile. The IBM PC/XT in 1983 included an internal 10 MB HDD, and soon afterwards the internal HDDs mushroomed on a personal computer.
External HDDs stay popular for longer in Apple Macintosh. Many Macintosh computers created between 1986 and 1998 display the SCSI port on the back, making external expansion simpler. Older older Macintosh computers do not have user-accessible hard drive space (of course, Macintosh 128K, Macintosh 512K, and Macintosh Plus do not have hard drive spots at all), so on those models the external SCSI disk is the only option that it makes sense to expand any internal storage.
Thailand's floods in 2011 damaged the plant and impacted on the cost of hard disk drives that were disadvantageous between 2011 and 2013.
Driven by increased area density, HDD continues to be improved; some highlights are listed in the table above. Market applications flourished through the 2000s, from mainframe computers in the late 1950s to mass storage applications including computers and consumer applications such as entertainment content storage.
NAND performance increases faster than HDD, and applications for HDD experience erosion. The smaller form factor, 1.8-inches and below it, was discontinued around 2010. Solid-state storage prices (NAND), represented by Moore's law, are increasing faster than HDD. NAND has a higher demand price elasticity than HDD, and this drives market growth. During the late 2000s and 2010s, the life cycle of HDD products entered a mature phase, and sluggish sales could indicate the beginnings of a declining phase.
Maps Hard disk drive
Technology
Magnetic recording
Modern HDD record data with thin film magnetization of ferromagnetic materials on disk. Sequential changes in the direction of magnetization are binary data bits. Data is read from disk by detecting transitions in magnetization. User data is encoded using an encoding scheme, such as long run-length encoding, which determines how data is represented by magnetic transitions.
Typical HDD designs consist of spindle spindle spindles that store flat circular disks, also called disks, that hold down data recorded. The plates are made of non-magnetic materials, usually aluminum alloys, glass, or ceramics, and are coated with a shallow layer of magnetic material typically 10-20 nm in depth, with an outer layer of carbon for protection. For reference, a standard sheet of photocopy paper is 0.07-0.18 millimeters (70,000-180,000 nm).
The contemporary HDD disks rotate at a speed varying from 4,200 rpm in energy-efficient portable devices, up to 15,000 rpm for high-performance servers. The first HDD rotates at 1,200 rpm and, over the years, 3,600 rpm is the norm. In December 2013, the discs in most consumer-grade HDDs spin at 5,400 rpm or 7,200 rpm.
Information is written and read from a plate when it rotates through devices called read-and-write heads that are positioned to operate very close to the magnetic surface, with their flying altitude often in the range of tens of nanometers. The read-and-write head is used to detect and modify the magnetization of the material passing just below it.
On modern drives, there is one head for each surface of the magnetic plate on the spindle, which is mounted on the same arm. An actuator arm (or access arm) moves the head in an arc (roughly radial) across the disk as it spins, allowing each head to access almost the entire plate surface while rotating. The arm is moved using a voice coil actuator or in some older design stepper motor. The initial hard disk drive writes data at a few bits constant per second, so all tracks have the same amount of data per track but modern drives (since the 1990s) use zone bit tape - increase write speed from the inside to the outer zone and thus save more lots of data per track in the outer zone.
In modern drives, the small size of the magnetic region creates the danger that their magnetic state may be lost due to thermal effects, thermally induced magnetic instability commonly known as the "superparamagnetic limit". To overcome this, the disk is lined with two parallel magnetic layers, separated by a 3-atom layer of non-magnetic element ruthenium, and two magnetic layers in opposite orientation, thus reinforcing each other. Another technology used to overcome the thermal effects to allow greater recording density is perpendicular recording, first shipped in 2005, and in 2007 this technology was used in many HDDs.
In 2004, a new concept was introduced to allow for a further increase of data density in magnetic recording, using recording media consisting of soft and hard magnetic coatings combined. The so-called currency exchange media , also known as combined merge media exchange , allows good writing because of the soft-write help nature. However, thermal stability is determined only by the most difficult layers and is not affected by the soft layers.
Components
The typical HDD has two electric motors; spindle motors that play disks and actuators (motors) that position the read/write head assembly across the rotating disk. The disk motor has an external rotor attached to the disk; the stator winding stays in place. Across the actuator at the end of the supporting arm is the read-write head; thin thin-circuit cables connect the read-write head to the amplifier electronics mounted on the actuator shaft. The supporting arm of the head is very light, but also stiff; in modern drives, acceleration in the head reaches 550 g .
Akt actoror is a permanent magnet and a moving coil motor that swings the head to the desired position. The metal plate supports a high neodymium-iron-boron (NIB) magnesium magnet. Below this plate is a moving coil, often referred to as a voice coil with a coil analog on the loudspeaker, attached to the actuator hub, and below it is a second NIB magnet, mounted on the underside of the motor (some drive only has one magnet).
The sound coil itself is shaped somewhat like an arrowhead, and is made of a double-layered copper magnet wire. The inner layer is insulation, and the outside is a thermoplastic, which binds the coil together after it injures the shape, making it self-sufficient. The coil parts along both sides of the arrow head (which point to the center of the actuator bearing) then interact with the magnetic field of the fixed magnet. A current flowing radially along one side of the arrow and radially inward on the other produces a tangential force. If the magnetic field is uniform, each side will produce opposite forces that will cancel each other. Therefore, the magnetic surface is the northern half and the south pole, with the radial dividing line in the middle, causing the two sides of the coil to see the opposing magnetic field and produce an added force instead of canceling. The current along the top and bottom of the coil produces a radial force that does not turn the head.
The HDD electronics controls the actuator movement and disk rotation, and performs the on-demand reading and writing of the disk controller. Feedback from the drive electronics is achieved by means of a special segment of disk dedicated to servo feedback. This is a complete concentric circle (in the case of special servo technology), or segments interspersed with real data (in case of embedded servo technology). The servo feedback optimizes the signal to noise ratio of the GMR sensor by adjusting the voice-coil of the driven arm. Spinning disks also use servo motors. Modern disk firms are able to schedule efficient reading and writing on the platter surface and remapping the failed media sectors.
Error level and handling
Modern drives use extension error codes (ECCs) extensively, especially Reed-Solomon error correction. These techniques store extra bits, determined by mathematical formulas, for each data block; extra bits allow many errors to be corrected without being seen. The extra bit itself takes up space in the HDD, but allows higher recording density to be used without causing irreversible errors, resulting in much larger storage capacity. For example, a typical 1 TB hard drive with a 512-byte sector provides an additional capacity of about 93 GB for ECC data.
In the latest drive, in 2009, the low density parity check code (LDPC) replaced Reed-Solomon; The LDPC code allows performance that is close to the Shannon Limit and thus provides the highest available storage density.
The typical hard disk drive attempts to "re-map" the data in the failed physical sector to the physical sector of the backup provided by the "reserve pool" drive (also called the "reserve pool"), while relying on ECC to recover the stored data while the number of errors in the bad sector is still quite low. The SMART (Self-Monitoring, Analysis, and Reporting Technology) feature counts the total number of errors across HDD set by ECC (though not on all hard drives because the SMART attribute associated with "ECC Hardware Recoverable" and "Soft ECC Correction" is not supported consistent), and the total number of sector remotations performed, as the emergence of many of these errors can predict HDD failure.
The "No-ID Format", developed by IBM in the mid-1990s, contains information about which sectors are bad and where reconstructed sectors have been found.
Only a small part of the detected error ends up as irreparable. For example, the specification for an enterprise SAS disk (model from 2013) estimates this fraction to be an unrepaired error in every 10 16 bits, and another SAS enterprise disk from 2013 establishes a similar error rate. Another modern (per 2013) SATA disk company determines an error rate of less than 10 unrecoverable read errors in every 10 16 bits. A company disk with Fiber Channel interface, which uses 520 byte sectors to support the Data Integrity Field standard to combat data corruption, sets a similar error rate in 2005.
The worst type of error is a data corruption that is an error not detected by the disk firmware or host operating system; some of these errors may be caused by hard disk drive malfunctions.
Development
The rate of density development of the area is similar to Moore's law (doubling every two years) to 2010: 60% per annum during 1988-1996, 100% during 1996-2003 and 30% during 2003-2010. Gordon Moore (1997) mentions an increase in "flabbergasting," while observing later that growth can not continue forever. Price increases are slowing to -12% per year during 2010-2017, as growth in the area density slows. The rate of progress for area densities slows to 10% per year during 2010-2016, and there is difficulty in migrating from recording perpendicular to newer technology.
When the bit cell size decreases, more data can be inserted into one drive plate. In 2013, the 3 TB HDD desktop desktop (with four platter) will have a density of around 500 Gbit/in 2 which will amount to a bit cell consisting of about 18 magnetic grains (11 by 1.6 grains). Since the mid-2000s, increasing density areas have been increasingly challenged by superparamagnetic trilemma involving grain size, magnetic power of grain, and the ability to write heads. To maintain the received signal against noise, smaller grains are required; smaller grains can be turned on themselves (electrothermal instability) unless their magnetic strength is increased, but the known writing head material can not produce enough magnetic fields to write the medium. Several new magnetic storage technologies are being developed to address or at least reduce this trilemma and thus maintain HDD competitiveness with respect to products such as flash memory-based SSDs.
In 2013, Seagate introduced one such technology, a shingled magnetic recording (SMR). In addition, SMR comes with design complexity which can lead to reduced write performance. Other new recording technologies, in 2016, are still in development including heat-assisted magnetic recording (HAMR), microwaved magnetic recording (MAMR), two-dimensional magnetic recording (TDMR), bit patterned recordings (BPR), and heads giant magnetoresistance (CPP/GMR) that is perpendicular to the plane.
The growth rate of the area density has fallen below Moore's legal level by 40% per annum, and the deceleration is expected to last up to at least 2020. Depending on the feasibility and timing assumption of this technology, the median estimate by industry observers and analysts for 2020 and beyond for density growth the area is 20% per year with a range of 10-30%. The limits that can be achieved for HAMR technology in combination with BPR and SMR may be 10A, Tbit/in 2 , which will be 20 times higher than 500 Gbit/in 2 represented by HDD desktop production 2013. By 2015, HDD HAMR has been delayed for several years, and is expected by 2018. They require different architectures, with new media redesign and read/write heads, new lasers, and new optical transducers near the field.
Capacity
Hard disk drive capacity, as reported by the operating system to the end user, is smaller than the number declared by the manufacturer for several reasons: the operating system uses some space, the use of some space for data redundancy, and the use of space for the file system structure. Also, differences in reported capacity in prefix SI units versus binary prefixes can cause a false impression of lost capacity.
Calculation
Modern hard disk drives appear to their host controllers as a series of adjacent logical blocks, and the capacity of a dirty hard disk is calculated by multiplying the number of blocks by block size. This information is available from the manufacturer's product specifications, and from the drive itself through the use of operating system functions that invoke low-level drive commands.
The gross capacity of older HDDs is calculated as the product of the number of cylinders per recording zone, number of bytes per sector (most often 512), and number of drive zones. Some modern SATA drives also report cylinder-head-sector (CHS) capacity, but these are not physical parameters because the reported values ​​are limited by the historic operating system interface. The C/H/S scheme has been replaced by a logical addressing block (LBA), a simple linear addressing scheme that places blocks by an integer index, which starts in LBA 0 for the first block and additions thereafter. When using the C/H/S method to describe modern large drives, the number of heads is often set to 64, although typical hard disk drives, by 2013, have between one and four platter.
In modern HDDs, the reserve capacity for defective management is not included in the published capacity; however, in many early HDDs, certain sectors are reserved as spare parts, thereby reducing the capacity available to the operating system.
For RAID subsystems, data integrity and fault tolerance requirements also reduce the realized capacity. For example, a RAID 1 array has about half the total capacity as a result of data mirroring, while RAIDÃ,5 arrays with x drives lose 1/x of capacity (which is equal to capacity of one drive) because it stores parity information. The RAID subsystem is multiple drives that appear as one drive or more drive to the user, but provide fault tolerance. Most RAID vendors use checksums to improve data integrity at block level. Some vendors design systems using a HDD with a 520 byte sector to load 512 bytes of user data and eight byte checksums, or by using a separate 512-byte sector for checksum data.
Some systems may use hidden partitions for system recovery, reducing the capacity available to end users.
Format
Data is stored on the hard drive in a series of logical blocks. Each block is limited by markers identifying start and end, error detecting and correcting information, and space between blocks to allow for small time variations. This block often contains 512 bytes of data that can be used, but other sizes have been used. As drive density increases, an initiative known as Advanced Format extends block size to 4096 bytes of usable data, with significant reductions in the amount of disk space used for block headers, error checking data, and distances.
The process of initializing this logical block on a physical disk disk is called low-level formatting , which is usually done in the factory and is usually unchanged in the field. High-level formatting writes the data structures used by the operating system to organize the data files on the disk. This includes writing partition and file system structure to the selected logical block. For example, some disk space will be used to store the disk file name directory and a list of logical blocks associated with a particular file.
Examples of partition mapping schemes include Master boot record (MBR) and GUID Partition Table (GPT). Examples of data structures stored on the disk to retrieve files include File Allocation Table (FAT) in the DOS file system and inodes in many UNIX file systems, as well as other operating system data structures (also known as metadata). As a result, not all space on the HDD is available for user files, but the system overhead is usually small compared to user data.
Unit
Total HDD capacity is given by manufacturers using SI decimal prefixes such as gigabytes (1 GB = 1,000,000,000 bytes) and terabytes (1 TB = 1,000,000,000,000 bytes). This practice originated from the early days of computing; in the 1970s, "million", "mega" and "M" were consistently used in the decimal sense for drive capacity. However, memory capacity is quoted using a binary interpretation of the prefix, which uses 1024 instead of 1000 powers.
The software reports hard disk drives or memory capacity in various forms using decimal or binary prefixes. The Microsoft Windows family of operating systems use binary conventions when reporting storage capacity, so the HDD offered by the manufacturer as a 1 TB drive is reported by the operating system as 931 GB HDD. Mac OS X 10.6 ("Snow Leopard") uses decimal conventions when reporting HDD capacity. The default behavior of the df command line utility in Linux is reporting the capacity of the HDD as the number of 1024-byte units.
The difference between decimal and binary prefix interpretations causes some consumer confusion and causes a class action lawsuit against HDD manufacturers. The plaintiffs argue that the use of decimal prefixes effectively misleads consumers while defendants deny wrongdoing or obligations, insisting that their marketing and advertising comply with all things by law and that no class members are damaged or injured.
Evolution of prices
The price of HDD per byte increased at -40% per annum during 1988-1996, -51% per year during 1996-2003, and -34% per year during 2003-2010. The price increase slowed to -13% per year during 2011-2014, as increasing density of the area slowed and the flood of Thailand 2011 damaged the manufacturing facility.
Form factor
IBM's first hard drive, the IBM 350, uses a stack of fifty 24-inch disks and has a size comparable to two large refrigerators. In 1962, IBM introduced a 1311 disk model, which used six 14-inch (nominal) discs in removable packaging and about the size of a washing machine. It became the standard platter size and pushed form factor for years, which is also used by other manufacturers. The IBM 2314 uses a disc of the same size in eleven high packs and introduces the "drive in a drawer" layout, even though the "drawer" is not a complete drive.
Then the drive is designed to be fully chassis to be mounted on a 19-inch rack. RK05 and RL01 digital are early examples using a 14-inch single disc in detachable packaging, all mounting drives in a 10.5-inch (six rack) rack space. In the mid to late 1980s, the same-sized Fujitsu Eagle, which used a 10.5-inch dish, was by chance a popular product.
Such large disks have never been used with microprocessor-based systems. With the increasing sales of microcomputers having built-in floppy-disk drives (FDD), HDDs that will be suitable for FDD installation are desirable. Thus the HDD Form Factor , initially follows an 8-inch, 5.25-inch, and 3.5-inch floppy disk drive. Although referred to by this nominal size, the actual size for the three drives is 9.5 ", 5.75" and 4 "wide, respectively. Since there is no smaller floppy disk disk, smaller form factor HDDs developed from product or standard offerings industrial. inch drives are actually 2.75 "wide.
In 2012, 2.5 inches and 3.5 inches hard disk is the most popular measure. In 2009, all manufacturers have halted the development of new products for the form factor of 1.3 inches, 1 inch and 0.85 inches due to the decline in the price of flash memory, which has no moving parts. While the nominal size is in inches, the actual dimension is determined in millimeters.
Performance characteristics
Factors that limit the time to access data on HDDs are mostly related to the mechanical properties of rotating disks and moving heads.
Find time is a measure of how long the assembly head to travel to the disk path containing the data. The first HDD has an average time of about 600 ms; Some early PC drives used stepper motors to move their heads, and as a result had a chance to find 80-120 ms at a time, but this was quickly enhanced by the actuation of voice coil type in the 1980s, reducing the search time to about 20 ms. Look for time keeps increasing slowly over time. The fastest server drive currently has a search time of about 4 ms. The average search time is the right time to do all possible search divided by the sum of all possible searches, but in practice is determined by statistical methods or is only estimated as the time of searching for more than one-third of the number of tracks.
Rotation latency occurs because the desired disk sector may not be directly under the head when data transfer is requested. The average rotation rotation is shown in the table, based on the statistical relation that the average latency is one half of the rotation period.
The bit rate or data transfer rate (after the head is in the right position) creates a delay which is a function of the number of blocks being transferred; usually relatively small, but can be very long with large adjacent file transfers.
The delay can also occur if the disc drive is stopped to save energy.
Defragmentation is a procedure used to minimize delays in retrieving data by moving related items to the physical proximity area on the disk. Some computer operating systems do defragmentation automatically. Although automated defragmentation is intended to reduce access delays, performance will decrease temporarily while the procedure is in progress.
The time to access the data can be increased by increasing the rotation speed (thus reducing latency) or by reducing the time spent searching. Increased area density increases throughput by increasing data rates and by increasing the amount of data below a set of heads, thus potentially reducing search activity for a given amount of data. The time to access the data does not follow the increase in throughput, which can not keep up with the growth in bit density and storage capacity.
Latency
Data transfer rate
In 2010, the common 7200-rpm HDD desktop had a continuous "disk-to-buffer" data transfer rate of up to 1,030 Mbit/s. This level depends on the location of the track; this figure is higher for data on the outer tracks (where there are more data sectors per rotation) and lower towards deep tracks (where there are fewer data sectors per rotation); and generally somewhat higher for a 10,000-rpm drive. The currently widely used standard for buffer-to-computer interface is 3.0 Gbit/s SATA, which can send about 300 megabytes/s (10-bit encoding) from buffer to computer, and thus still comfortable in front of transfer rate disk-to-buffer today. The speed of data transfer (read/write) can be measured by writing large files to disk using a special file creation tool, then re-reading the file. Transfer rates can be affected by file system fragmentation and file layout.
The HDD data transfer rate depends on the speed of the disk rotation and the data recording density. Because heat and vibration limit the speed of rotation, advancing density becomes the main method for increasing sequential transfer rates. Higher speeds require a stronger spindle motor, which creates more heat. While increasing the area density by increasing both the number of tracks across the disk and the number of sectors per track, only the latter increases the rate of data transfer for the given rpm. Since data transfer rates only track one of the two component density areas, their performance increases at a lower rate.
Other considerations
Other performance considerations include price adjusted for quality, power consumption, audible noise, and both operating and non-operating shock resistance.
The Federal Reserve Board has a quality-adjusted price index for large-scale enterprise storage systems including three or more corporate HDDs and associated controls, racks, and cables. Prices for this large-scale storage system increased at -30% per year during 2004-2009 and -22% per year during 2009-2014.
Access and interface
The hard drive is currently connected to the computer through one of several bus types, including parallel ATA, Serial ATA, SCSI, Serial Attached SCSI (SAS), and Fiber Channel. Some drives, especially external portable drives, use IEEE 1394, or USB. All of these interfaces are digital; The electronics on the drive process analog signals from the read/write head. The current drive presents a consistent interface to the entire computer, regardless of the data encoding scheme used internally, and is independent of the physical amount of disks and heads in the drive.
Typically DSP in electronics inside the drive takes raw analog voltage from the read head and uses PRML and Reed-Solomon error correction to decode the data, then sends the data out of the standard interface. The DSP also sees the level of error detected by error detection and correction, and performs poor sector mapping, data collection for Self-Monitoring, Analysis, and Reporting Technology, and other internal tasks.
The modern interface connects the drive to the host interface with a single data/control cable. Each drive also has an additional power cord, usually directly to the power supply unit. The older interface has separate cables for data signals and for drive control signals. Small Computer System Interface (SCSI), originally called SASI for Shugart Associates System Interface, was standard on servers, workstations, Commodore Amiga, Atari ST and Apple Macintosh computers until the mid-1990s, Large models have been routed to IDE (and later, SATA) family disks. The data cable length limit allows external SCSI devices.
Integrity and failure
Due to the very close distance between the head and the surface of the disk, the HDD is vulnerable to damage by head crash - a disk failure where the head scratches the surface of the platter, frequently grinding the thin magnetic film and causing data loss. Head damage can be caused by electronic failure, sudden power failure, physical shock, contamination of internal drive cover, wear, corrosion, or bad grip and head.
The HDD spindle system relies on the air density within the scope of the disk to support the head at the correct flying height when the disk is rotating. HDD requires a certain air density range to operate properly. Connection to the external environment and density occurs through a small hole in the enclosure (about 0.5 mm in width), usually with an inner filter ( breath filter ). If the air density is too low, then there is not enough lift for the flying head, so the head is too close to the disk, and there is a risk of crash heads and data loss. Special manufactured sealed and pressurized disks are required for reliable high altitude operations, above about 3,000 m (9,800 ft). Modern disks include temperature sensors and adjust their operation to the operating environment. The resting hole can be seen on all disk drives - they usually have a sticker next to it, warning the user not to close the hole. The air inside the drive operation also kept moving, swept away by friction with a rotating disk. This air passes through an internal recirculating filter (or "recirculation") to remove any residual contaminants from any particles, particles or chemicals that have somehow entered the enclosure, and internally generated particles or outgassing in normal operation. High humidity for a long time can cause corrosion of the head and disk.
For magnetoresistive giant heads (GMR) in particular, minor damage to the head due to contamination (which does not remove the magnetic surface of the disk) still causes the head to suffer from temporary heat, due to friction with the surface of the disk, and may cause the data to be unreadable. for a short time until a stable head temperature (called "thermal asperity", a problem that can be partially handled by precise electronic filtering of the read signal).
When the hard disk logic board fails, the drive can often be restored to a functioning sequence and data recovered by replacing the circuit board with one of the identical hard disks. In the case of read-write head errors, they can be replaced using special tools in a dust-free environment. If the disk disks are not damaged, they can be transferred into the same enclosure and the data can be copied or cloned to a new drive. If a disk failure occurs, disconnect and disk disc imaging may be required. For logical damage to the file system, various tools, including fsck on UNIX-like systems and CHKDSK in Windows, can be used for data recovery. Recovery from logical damage can require carving files.
The general expectation is that hard disk drives designed and marketed for server use will fail less often than consumer-level drives typically used on desktop computers. However, two independent studies by Carnegie Mellon University and Google found that the "level" of a drive was not related to the drive failure rate.
Summary of 2011 research into the SSD and magnetic disk failure patterns by Tom Hardware summarizes the research findings as follows:
- Means the time between failure (MTBF) does not show reliability; annual failure rate is higher and usually more relevant.
- Magnetic disks have no specific tendency to fail during initial use, and temperatures have little effect; on the contrary, failure rates continue to increase with age.
- S.M.A.R.T. warn of mechanical problems but no other problems that affect reliability, and therefore not a reliable condition indicator.
- Drive failure rates sold as "corporate" and "consumer" are "very similar", although these types of drives are tailored for their different operating environments.
- In the drive array, a single drive failure significantly increases the short-term risk of a second drive failure.
Market segment
- Desktop HDD
- They typically store between 60 GB and 4 TB and rotate at 5,400 to 10,000 rpm, and have a media transfer rate of 0.5 Gbit/s or higher (1 GB = 10 9 bytes; 1 Gbit/s = 10 9 bit/s). In February 2017, the highest-capacity HDD desktop stores 12 TB, with plans to release a 14TB one in 2017. By 2016, the typical hard drive speed on average desktop computers is 7200 RPM, while low-cost desktop computers can use 5900 RPM or 5400 RPM drive. For some time in the 2000s and early 2010s some desktop users will also be using 10k RPM drives such as Western Digital Raptor but those drives have become less frequent by 2016 and are not commonly used now, have been replaced by NAND flash based SSDs.
- Mobile (laptop) HDD
- Smaller than their desktop counterparts and companies, they tend to be slower and have lower capacity. Mobile HDD rotates at 4,200 rpm, 5,200 rpm, 5,400 rpm, or 7,200 rpm, with a typical 5,400 rpm. 7,200 rpm drives tend to be more expensive and have smaller capacity, while the 4,200 rpm model usually has a very high storage capacity. Because of the smaller plate (s), mobile HDDs generally have lower capacities than their desktop counterparts.
- There is also a 2.5-inch drive that spins at 10,000 rpm, which is included in the enterprise segment with no intention of being used in laptops.
- Enterprise HDD
- Usually used with many user computers running enterprise software. Examples are: transaction processing database, internet infrastructure (email, webserver, e-commerce), scientific computing software, and nearline storage management software. Enterprise drives generally operate continuously ("24/7") in demanding environments while delivering the highest performance without sacrificing reliability. Maximum capacity is not the ultimate goal, and as a result drives are often offered in relatively low capacity in relation to their costs.
- The fastest enterprise HDD spins at 10,000 or 15,000 rpm, and can achieve a sequential media transfer rate above 1.6 Gbit/s and a continuous transfer rate of up to 1 Gbit/s. Drives run at 10,000 or 15,000 rpm using smaller disks to reduce the need for increased power (since they have fewer air resistance) and therefore generally have lower capacities than the highest capacity desktop drives. Enterprise HDDs are usually connected via Serial Attached SCSI (SAS) or Fiber Channel (FC). Some support multiple ports, so they can connect to the redundant host bus adapter.
- HDD Enterprise can have sector sizes larger than 512 bytes (often 520, 524, 528 or 536 bytes). Additional space per sector can be used by hardware RAID controllers or applications to store Data Integrity Field (DIF) or Data Integrity Extensions (DIX) data, which results in higher reliability and quiet data corruption prevention.
- HDD consumer electronics
- They include drives embedded into digital video recorders and automotive vehicles. The first is configured to provide guaranteed streaming capacity, even in the face of read and write errors, while the latter is built to withstand a larger number of surprises. They usually spin at speeds of 5400 RPM.
Manufacturers and sales
More than 200 companies have been producing HDDs from time to time. But consolidation has concentrated production to only three manufacturers today: Western Digital, Seagate, and Toshiba.
Worldwide revenue for disk storage is down 4% annually, from a peak of $ 38 billion in 2012 to $ 27 billion by 2016. HDD production grows 16% annually, from 335 exabytes in 2011 to 693 exabytes by 2016. Shipments are down 7% annually over this time period, from 620 million units to 425 million. Seagate and Western Digital each have 40-45% of the unit shipments, while Toshiba has 13-17%. The average selling price for the two largest manufacturers is $ 60 per unit by 2015.
Competition from solid-state drive
The maximum storage density for flash memory used in solid state drives (SSDs) is 2.8Ã, Tbit/in 2 in a laboratory demonstration in 2016, and the maximum for HDD is 1.5Ã, Tbit/in < soup> 2 . The area of ​​flash memory density doubled every two years, similar to Moore's law (40% per year) and faster than 10-20% per year for HDD. In 2016, the maximum capacity is 10 terabytes for HDD, and 15 terabytes for SSD. HDD is used on 70% of desktop and notebook computers manufactured in 2016, and SSDs are used in 30%. The share of HDD usage is decreasing and could fall below 50% in 2018-2019 according to one estimate, since SSDs replace smaller capacities (less than one terabyte) of HDDs on desktop and notebook computers and MP3 players.
The market for silicon-based flash memory chips (NAND), used in SSDs and other applications, is growing rapidly. World revenue grew 12% per year during 2011-2016. This increased from $ 22 billion in 2011 to $ 39 billion in 2016, while production grew 46% annually from 19 exabytes to 120 exabytes.
External hard disk drive
External hard disks are usually connected via USB; variants using the USB 2.0 interface generally have a slower data transfer rate when compared to an internal installed hard drive connected via SATA. The plug and play drive functionality offers system compatibility and features large storage options and portable design. As of March 2015, the available capacity for external hard disks ranges from 500 GB to 10 TB.
External hard disk drives are usually available as pre-assembled integrated products, but can also be assembled by combining external enclosures (with USB or other interfaces) with drives purchased separately. They are available in sizes of 2.5 inches and 3.5 inches; The 2.5-inch variant is usually called the portable external drive , while the 3.5-inch variant is referred to as the desktop external drive . "Portable" drives are packed in smaller scope and lighter than "desktop" drives; In addition, the "portable" drives use the power provided by the USB connection, while the "desktop" drive requires an external power brick.
Features such as biometric security or multiple interfaces (eg, Firewire) are available at a higher cost. External hard disk drives are assembled earlier, which, when removed from their enclosures, can not be used internally on laptops or desktop computers because of the embedded USB interface on their printed circuit boards, and the lack of SATA (or Parallel ATA) interfaces.
See also
Note
References
Further reading
- Mueller, Scott (2011). Upgrade and Fix PC (20th edition). Que. ISBNÃ, 0-7897-4710-3.
- Messmer, Hans-Peter (2001). The PC Hardware Book is Not Replaceable (fourth edition). Addison-Wesley. ISBN: 0-201-59616-4.
External links
- Hard Disk Drives Encyclopedia
- The video displays the opened HD that works
- Average time looking for a computer disk
- Timeline: 50 Years Hard Drives
- HDD from inside: Tracks and Zones. How difficult is that?
- Hard disk hacking - firmware modifications, in eight sections, will extend as far as booting the Linux kernel on a regular HDD controller
- Hiding Data in Hard Drive Service Area, February 14, 2013, by Ariel Berkman
- Rotary Accelerated Feed Information Sheet (RAFF), Western Digital, January 2013
- PowerChoice Technology for Hard Disk Drive Power Savings and Flexibility, Seagate Technology, March 2010
- Shingled Magnetic Recording (SMR), HGST, Inc., 2015
- The road to Helium, HGST, Inc., 2015
- Research paper on the use of a magnetic photoconductor perspective in magneto-optical data storage.
Source of the article : Wikipedia
0 Comments