青春的岁月，我们身不由己: ATA-Advanced Technology Attachment

Advanced Technology Attachment (ATA) is a standard interface for connecting storage devices such as hard disks and CD-ROM drives inside personal computers.

The standard is maintained by X3/INCITS committee T13. Many synonyms and near-synonyms for ATA exist, including abbreviations such as IDE and ATAPI. Also, with the market introduction of Serial ATA in 2003, the original ATA was retroactively renamed Parallel ATA (PATA).

In line with the original naming, this article covers only Parallel ATA.

Parallel ATA standards allow cable lengths up to only 18 inches (46 centimetres) although cables up to 36 inches (91 cm) can be readily purchased. Because of this length limit, the technology normally appears as an internal computer storage interface. It provides the most common and the least expensive interface for this application.

[hide]

[edit] History

ATA connection sockets on a PC motherboard located below RAM sockets

The name of the standard was originally conceived as PC/AT Attachment as its primary feature was a direct connection to the 16-bit ISA bus then known as 'AT bus'; the name was shortened to an inconclusive "AT Attachment" to avoid possible trademark issues.

An early version of the specification, conceived by Western Digital in late 1980s, was commonly known as Integrated Drive Electronics (IDE) due to the drive controller being contained on the drive itself as opposed to the then-common configuration of a separate controller connected to the computer's motherboard — thus making the interface on the motherboard a host adapter, though many people continue, by habit, to call it a controller.

Enhanced IDE (EIDE) — an extension to the original ATA standard again developed by Western Digital — allowed the support of drives having a storage capacity larger than 528 megabytes (504 mebibytes), up to 8.4 gigabytes. Although these new names originated in branding convention and not as an official standard, the terms IDE and EIDE often appear as if interchangeable with ATA. This may be attributed to the two technologies being introduced with the same consumable devices — these "new" ATA hard drives.

With the introduction of Serial ATA around 2003, conventional ATA was retroactively renamed to Parallel ATA (P-ATA), referring to the method in which data travels over wires in this interface.

The interface at first worked only with hard disks, but eventually an extended standard came to work with a variety of other devices — generally those using removable media. Principally, these devices include CD-ROM and DVD-ROM drives, tape drives, and large-capacity floppy drives such as the Zip drive and SuperDisk drive. The extension bears the name AT Attachment Packet Interface (ATAPI), which started as non-ANSI SFF-8020 standard developed by Western Digital and Oak Technologies, but then included in the full standard now known as ATA/ATAPI starting with version 4. Removable media devices other than CD and DVD drives are classified as ARMD (ATAPI Removable Media Device) and can appear as either a floppy or a hard drive to the operating system.

The move from programmed input/output (PIO) to direct memory access (DMA) provided another important transition in the history of ATA. As every computer word must be read by the CPU individually, PIO tends to be slow and use a lot of CPU resources. This is especially a problem on faster CPUs where accessing an address outside of the cacheable main memory (whether in the I/O map or the memory map) is a relatively expensive process. This meant that systems based around ATA devices generally performed disk-related activities much more slowly than computers using SCSI or other interfaces. However, DMA (and later Ultra DMA, or UDMA) greatly reduced the amount of processing time the CPU had to use in order to read and write the disks. This is possible because DMA and UDMA allow the disk controller to write data to memory directly, thus bypassing the CPU.

The original ATA specification used a 28-bit addressing mode. This allowed for the addressing of 2²⁸ (268,435,456) sectors (with blocks of 512 bytes each), resulting in a maximum capacity of 137 gigabytes (128 GiB). The standard PC BIOS system supported up to 7.88 GiB (8.46 GB), with a maximum of 1024 cylinders, 256 heads and 63 sectors. When the lowest common denominators of the CHS limitations in the standard PC BIOS system and the IDE standard were combined, the system as a whole was left limited to a mere 504 megabytes. BIOS translation and LBA were introduced, removing the need for the CHS structure on the drive itself to match that used by the BIOS and consequently allowing up to 7.88 GiB when accessed through Int 13h interface. This barrier was overcome with Int 13H extensions, which used 64 bit linear address and therefore allowed access to the full 128 GiB and more (although some BIOSes initially had problems handling more than 31.5 GiB due to a bug in implementation).

ATA-6 introduced 48 bit addressing, increasing the limit to 128 PiB (or 144 petabytes). Some OS environments, including Windows 2000 until Service Pack 3, did not enable 48-bit LBA by default, so the user was required to take extra steps to get full capacity on a 160 GB drive.

All these size limitations come about because some part of the system is unable to deal with block addresses above some limit. This problem may manifest itself by the system recognizing no more of a drive than that limiting value, or by the system refusing to boot and hanging on the BIOS screen at the point when drives are initialized. In some cases, a BIOS upgrade for the motherboard will resolve the problem. This problem is also found in older external FireWire disk enclosures, which limit the usable size of a disk to 128 GB. By early 2005 most enclosures available have practically no limit. (Earlier versions of the popular Oxford 911 FireWire chipset had this problem. Later Oxford 911 versions and all Oxford 922 chips resolve the problem.)

[edit] Parallel ATA interface

Until the introduction of Serial ATA, 40-pin connectors generally attached drives to a ribbon cable. Each cable has two or three connectors, one of which plugs into an adapter that interfaces with the rest of the computer system. The remaining one or two connectors plug into drives. Parallel ATA cables transfer data 16 bits at a time (it is a common misconception that they transfer 32 bits of data at a time, mainly because the 40 cable ribbon would appear to allow this).

Parallel ATA Pins
Pin	Function	Pin	Function
1	Reset	2	Ground
3	Data 7	4	Data 8
5	Data 6	6	Data 9
7	Data 5	8	Data 10
9	Data 4	10	Data 11
11	Data 3	12	Data 12
13	Data 2	14	Data 13
15	Data 1	16	Data 14
17	Data 0	18	Data 15
19	Ground	20	Key (alternative usage is VCC_in)
21	DDRQ	22	Ground
23	I/O Write	24	Ground
25	I/O Read	26	Ground
27	IOC HRDY	28	Cable Select (see below)
29	DDACK	30	Ground
31	IRQ	32	No Connect
33	Addr 1	34	GPIO_DMA66_Detect (see below)
35	Addr 0	36	Addr 2
37	Chip Select 1P	38	Chip Select 3P
39	Activity	40	Ground

ATA's ribbon cables had 40 wires for most of its history, but an 80-wire version appeared with the introduction of the Ultra DMA/66 (UDMA4) mode. All of the additional wires in the new cable are ground wires, interleaved with the previously defined wires. The interleaved ground wire reduces the effects of capacitive coupling between neighboring signal wires, thereby reducing crosstalk. Capacitive coupling is more of a problem at higher transfer rates, and this change was necessary to enable the 66 megabytes per second (MB/s) transfer rate of UDMA4 to work reliably. The faster UDMA5 and UDMA6 modes also require 80-conductor cables.

Though the number of wires doubled, the number of connector pins and the pinout remain the same as on 40-conductor cables, and the external appearance of the connectors is identical. Internally, of course, the connectors are different: The connectors for the 80-wire cable connect a larger number of ground wires to a smaller number of ground pins, while the connectors for the 40-wire cable connect ground wires to ground pins one-for-one. 80-wire cables usually come with three differently colored connectors (blue, gray & black) as opposed to uniformly colored 40-wire cable's connectors (all black). The gray connector has pin 28 CSEL not connected; this makes it the slave position for drives configured cable select.

[edit] Using non-standard cables

The ATA standard has always specified a maximum cable length of just 46 cm (18 inches) and flat cables with particular impedance and capacitance characteristics. For various reasons, it may be desirable to use alternative cables : eg to have longer cables when connecting drives within a large computer case, or when mounting several physical drives into one computer ; or to use rounded cables to improve airflow (cooling) inside the computer case. Such cables are widely available on the market, and used successfully in most cases, however the user must understand that they are outside the parameters set by the specifications, and should be used with caution.

The short standard cable length all but completely eliminates the possibility of using parallel ATA for external devices.

[edit] Pin 20

In the ATA standard, Pin 20 is defined as key and is not used. However, some FLASH disks can use pin 20 as VCC_in to power disk without need of special power cable[1].

[edit] Pin 28

Pin 28 of the gray connector of an 80 conductor cable is not attached to any conductor of the cable. It is attached normally on the blue and black connectors.

[edit] Pin 34

Pin 34 is connected to ground inside the blue connector of an 80 conductor cable but not attached to any conductor of the cable. It is attached normally on the gray and black connectors. See page 315 of [2].

[edit] Differences between connectors on 80 conductor cables

The image shows PATA connectors after removal of strain relief, cover, and cable. Pin one is at bottom left of the connectors, pin 2 is top left, etc., except that the lower image of the blue connector shows the view from the opposite side, and pin one is at top right.

Each contact comprises a pair of points which together pierce the insulation of the ribbon cable with such precision that they make an excellent connection to the desired conductor without harming the insulation on the neighboring wires. The center row of contacts are all connected to the common ground bus and attach to the odd numbered conductors of the cable. The top row of contacts are the even-numbered sockets of the connector (mating with the even-numbered pins of the receptacle) and attach to every other even-numbered conductor of the cable. The bottom row of contacts are the odd-numbered sockets of the connector (mating with the odd-numbered pins of the receptacle) and attach to the remaining even-numbered conductors of the cable. (An alternate version of the connectors is allowed in which the even-numbered conductors are grounded and the odd-numbered conductors carry the signals. Obviously all three connectors on a cable must agree on this.)

Note the connections to the common ground bus from sockets 2 (top left), 19 (center bottom row), 22, 24, 26, 30, and 40 on all connectors. Also note (enlarged detail, bottom, looking from the opposite side of the connector) that socket 34 of the blue connector does not contact any conductor but unlike socket 34 of the other two connectors, it does connect to the common ground bus. On the gray connector, note that socket 28 is completely missing, so that pin 28 of the drive attached to the gray connector will be open. On the black connector, sockets 28 and 34 are completely normal, so that pins 28 and 34 of the drive attached to the black connector will be connected to the cable. Pin 28 of the black drive reaches pin 28 of the host receptacle but not pin 28 of the gray drive, while pin 34 of the black drive reaches pin 34 of the gray drive but not pin 34 of the host. Instead, pin 34 of the host is grounded.

The standard dictates color-coded connectors for easy identification by both installer and cable maker. All three connectors are different from one another. The blue (host) connector has the socket for pin 34 connected to ground inside the connector but not attached to any conductor of the cable. Since the old 40 conductor cables do not ground pin 34, the presence of a ground connection indicates that an 80 conductor cable is installed. The wire for pin 34 is attached normally on the other types and is not grounded. Installing the cable backwards (with the black connector on the system board, the blue connector on the remote device and the gray connector on the center device) will ground pin 34 of the remote device and connect host pin 34 through to pin 34 of the center device. The gray center connector omits the connection to pin 28 but connects pin 34 normally, while the black end connector connects both pins 28 and 34 normally.

(Although the standard itself must be purchased or consulted at a library, there is a draft copy available. See page 315 of the draft at [3], also archived at [4])

[edit] Multiple devices on a cable

If two devices attach to a single cable, one is commonly referred to as a master and the other as a slave. The master drive generally appears first when the computer's BIOS and/or operating system enumerates available drives. On old BIOSes (486 era and older) the drives are often misleadingly referred to by the BIOS as "C" for the master and "D" for the slave.

If there is a single device on a cable, in most cases it should be configured as master. However, some hard drives have a special setting called single for this configuration (Western Digital, in particular). Also, depending on the hardware and software available, a single drive on a cable can work reliably even though configured as the slave drive (this configuration is most often seen when a CDROM has a channel to itself).

[edit] Cable select

A drive setting called cable select was described as optional in ATA-1 and has come into fairly widespread use with ATA-5 and later. A drive set to "cable select" automatically configures itself as master or slave, according to its position on the cable. Cable select is controlled by pin 28. The host adapter grounds this pin; if a device sees that the pin is grounded, it becomes the master device; if it sees that pin 28 is open, the device becomes the slave device.

This setting is usually chosen by placing a jumper on the "cable select" position, usually marked CS, rather than on the "master" or "slave" position.

With the 40-wire cable it was very common to implement cable select by simply cutting the pin 28 wire between the two device connectors. This puts the slave device at the end of the cable, and the master on the "middle" connector. This arrangement eventually was standardized in later versions of the specification. If there is just one device on the cable, this results in an unused "stub" of cable. This is undesirable, both for physical convenience and electrical reasons: The stub causes signal reflections, particularly at higher transfer rates.

When the 80-wire cable was defined for use since ATAPI5/UDMA4, the master device goes at the end of the 18-inch cable(black connector), the middle-slave connector is grey, the blue connector goes onto the motherboard. So, if there is only one (master)device on the cable, there is no cable "stub" to cause reflections. Also, cable select is now implemented in the slave device connector, usually simply by omitting the ?contact? from the connector body. Both the 40-wire and 80-wire parallel-IDE cables share the same 40-socket connector configuration.

[edit] Master and slave clarification

Although they are in extremely common use, the terms master and slave do not actually appear in current versions of the ATA specifications. The two devices are correctly referred to as device 0 (master) and device 1 (slave), respectively. It is a common myth that "the master drive arbitrates access to devices on the channel" or that "the controller on the master drive also controls the slave drive." In fact, the drivers in the host operating system perform the necessary arbitration and serialization (as described in the next section), and each drive's controller operates independently. There is therefore no suggestion in the ATA protocols that one device has to ask the other if it can use the channel. Both are really "slaves" to the driver in the host OS.

[edit] Serialized, overlapped, and queued operations

The parallel ATA protocols up through ATA-3 require that once a command has been given to one device on an ATA interface, that command must complete before any subsequent command may be given to either device on the same interface. In other words, operations on the devices must be serialized—with only one operation in progress at a time—with respect to the ATA host interface.

For example, suppose a read operation is in progress on one drive on a given interface (cable). It is not possible to initiate another command on another drive on the same interface, or inform the first drive of additional commands that it should perform later (capabilities referred to as "overlapped" and "queued" operations, respectively), until the first drive's read operation completes. This is true even though the total I/O time is dominated by seek time and rotational latency, and during these phases, the first drive is transferring no data.

A useful mental model is that the host ATA interface is busy with the first request for its entire duration, and therefore can't be told about another request until the first one is complete.

The function of serializing requests to the interface is usually performed by a device driver in the host operating system.

The ATA-4 and subsequent versions of the specification have included both an "overlapped feature set" and a "queued feature set" as optional features. However, support for these is extremely rare in actual parallel ATA products and device drivers.

By contrast, overlapped and queued operations have been common in other storage buses for some time. In particular, tagged command queuing is characteristic of SCSI, and this has long been seen as a major advantage of SCSI over parallel ATA. The Serial ATA standard has supported what it calls native command queueing since its first released version, but the feature is present in only a few (generally the highest-priced) Serial ATA drives.

[edit] Two Devices on one cable - speed impact

There are many debates about how much a slow device can impact the performance of a faster device on the same cable. There is an effect, but the debate is confused by the blurring of two quite different causes, called here "Slowest Speed" and "One Operation at a Time".

[edit] "Slowest Speed"

It is a common misconception that, if two devices of different speed capabilities are on the same cable, both devices' data transfers will be constrained to the speed of the slower device.

For all modern ATA host adapters (since, at least, the late Pentium III and AMD K7 era) this is not true, as modern ATA host adapters support independent device timing. This allows each device on the cable to transfer data at its own best speed.

Even with older adapters without independent timing, this effect only impacts the data transfer phase of a read or write operation. This is usually the shortest part of a complete read or write operation (except for burst mode transfers).

[edit] "One Operation at a Time"

This is a much more important effect. It is caused by the omission of both overlapped and queued feature sets from most parallel ATA products. This means that only one device on a cable can perform a read or write operation at one time. Therefore, a fast device on the same cable as a slow device under heavy use will find that nearly every time it is asked to perform a transfer, it has to wait for the slow device to finish its own ponderous transfer.

For example, consider an optical device such as a DVD-ROM, and a hard drive on the same parallel ATA cable. With average seek and rotation speeds for such devices, a read operation to the DVD-ROM will take an average of around 100 milliseconds, while a typical fast parallel ATA hard drive can complete a read or write in less than 10 milliseconds. This means that the hard drive, if unencumbered, could perform more than 100 operations per second (and far more than that if only short head movements are involved). But since the devices are on the same cable, once a "read" command is given to the DVD-ROM, the hard drive will be inaccessible (and idle) for as long as it takes the DVD-ROM to complete its read—seek time included. Frequent accesses to the DVD-ROM will therefore vastly reduce the maximum throughput available from the hard drive. If the DVD-ROM is kept busy with average-duration requests, and if the host operating system driver sends commands to the two drives in a strict "round robin" fashion, then the hard drive will be limited to about 10 operations per second while the DVD-ROM is in use, even though the burst data transfers to and from the hard drive still happen at the hard drive's usual speed.

The impact of this on a system's performance depends on the application. For example, when copying data from an optical drive to a hard drive (such as during software installation), this effect probably doesn't matter: Such jobs are necessarily limited by the speed of the optical drive no matter where it is. But if the hard drive in question is also expected to provide good throughput for other tasks at the same time, it probably should not be on the same cable as the optical drive.

Remember that this effect occurs only if the slow drive is actually being accessed. The mere presence of an idle drive will not affect the performance of the other device on the cable (for a modern host adapter which supports independent timing).

[edit] ATA standards versions, transfer rates, and features

The following table shows the names of the versions of the ATA standards and the transfer modes and rates supported by each. Note that the transfer rate for each mode (for example, 66.7 MB/s for UDMA4, commonly called "Ultra-DMA 66") gives its maximum theoretical transfer rate on the cable. This is simply two bytes multiplied by the effective clock rate, and presumes that every clock cycle is used to transfer end-user data. In practice, of course, protocol overhead reduces this value.

Congestion on the host bus to which the ATA adapter is attached may also limit the maximum burst transfer rate. For example, the maximum data transfer rate for conventional PCI bus is 133 MB/s, and this is shared among all active devices on the bus.

In addition, no ATA hard drives exist capable of measured sustained transfer rates of above 80 MB/s. Furthermore, sustained transfer rate tests do not give realistic throughput expectations for most workloads: They use I/O loads specifically designed to encounter almost no delays from seek time or rotational latency. Hard drive performance under most workloads is limited first and second by those two factors; the transfer rate on the bus is a distant third in importance. Therefore, transfer speed limits above 66 MB/s really affect performance only when the hard drive can satisfy all I/O requests by reading from its internal cache — a very unusual situation, especially considering that such data is usually already buffered by the operating system.

Standard	Other Names	Transfer Modes Added (MB/s)	Maximum disk size	Other New Features	ANSI Reference
ATA-1	ATA, IDE	PIO 0, 1, 2 (3.3, 5.2, 8.3) Single-word DMA 0, 1 ,2 (2.1, 4.2, 8.3) Multi-word DMA 0 (4.2)	137 GB		X3.221-1994 (obsolete since 1999)
ATA-2	EIDE, Fast ATA, Fast IDE, Ultra ATA	PIO 3, 4: (11.1, 16.6) Multi-word DMA 1, 2 (13.3, 16.6)		28-bit logical block addressing (LBA)	X3.279-1996 (obsolete since 2001)
ATA-3	EIDE			S.M.A.R.T., Security	X3.298-1997 (obsolete since 2002)
ATA/ATAPI-4	ATA-4, Ultra ATA/33	Ultra DMA 0, 1, 2 (16.7, 25.0, 33.3) aka UDMA/33		AT Attachment Packet Interface (ATAPI), i.e. support for CD-ROM, tape drives etc., Optional overlapped and queued command set features, Host Protected Area (HPA)	NCITS 317-1998
ATA/ATAPI-5	ATA-5, Ultra ATA/66	Ultra DMA 3, 4 (44.4, 66.7) aka UDMA/66		80-wire cables	NCITS 340-2000
ATA/ATAPI-6	ATA-6, Ultra ATA/100	UDMA 5 (100) aka UDMA/100	144 PB	48-bit LBA, Device Configuration Overlay (DCO), Automatic Acoustic Management	NCITS 361-2002
ATA/ATAPI-7	ATA-7, Ultra ATA/133	UDMA 6 (133) aka UDMA/133 SATA/150		SATA 1.0, Streaming feature set, long logical/physical sector feature set for non-packet devices	NCITS 397-2005 (vol 1) NCITS 397-2005 (vol 2) NCITS 397-2005 (vol 3)
ATA/ATAPI-8	ATA-8	—		Hybrid drive featuring non-volatile cache to speed up critical OS files	In progress

In August 2004, Sam Hopkins and Brantley Coile of Coraid specified a lightweight ATA-over-Ethernet protocol to carry ATA commands over Ethernet instead of directly connecting them to a PATA host adapter. This permitted the established block protocol to be reused in Network-attached storage applications.

青春的岁月，我们身不由己

Monday, April 30, 2007

ATA-Advanced Technology Attachment

Contents