From Wikipedia, the free encyclopedia

This is a list of device bandwidths, or the physical layer information rates more properly known as net bit rates or 'useful' bit rates, at which digital devices can communicate over various kinds of buses and networks. The distinction between a bus (internal to a box and usually relying on many parallel wires) and a communications network cable (external, between boxes and rarely relying on more than four wires) can be arbitrary and many device interfaces or protocols (like SATA, USB, SCSI, PCI and a few variants of Ethernet) are used both inside a many-device box like a PC or one-device-box such as a hard drive enclosure. Accordingly this page lists both the internal ribbon and external communications cable standards together in one sortable table.

Factors limiting actual performance, criteria for real decisions

Most of the listed speeds are theoretical maximum throughput measures; in practice, the actual effective throughput are almost inevitably lower in proportion to the load from other devices (network/bus contention), interframe gap and other overhead in data link layer protocols, etc. The maximum goodput, for example the file transfer rate, may be even lower due to higher layer protocol overhead, and due to data packet retransmissions caused by line noise or interference such as crosstalk, or lost packets in congested intermediate network nodes. All protocols lose something, and the more robust ones that deal resiliently with very many failure situations tend to lose more maximum throughput to get higher total long term rates.

Device interfaces where one bus transfers data via another will be limited to the throughput of the slowest interface, at best. For instance SATA 6G controllers on one PCIe 5G channel will be limited to the 5G speed and have to employ more channels to get around this problem. Early implementations of new protocols very often have this kind of problem. The physical phenomena on which the device relies (such as spinning platters in a hard drive) will also impose limits, for instance no spinning platter shipping in 2009 saturates SATA3 so moving from this 3gbps interface to USB3 at 4.8gbps for one spinning drive will result in no increase in realized transfer rate. It might however be sensible for other reasons such as standardizing on a USB-only storage subsystem or exploiting USB's one-cable power.

Contention in a wireless or noisy spectrum where the physical medium is entirely out of the control of those who specify the protocol requires measures that also use up throughput. Wireless devices, BPL and modems may produce a higher line rate or gross bit rate, due to error-correcting codes and other physical layer overhead. It is extremely common for throughput to be far less than half of theoretical maximum though the more recent technologies (notably BPL) employ pre-emptive spectrum analysis to avoid this and so have much more potential to reach actual gigabit speeds in practice than prior modems.

Another factor reducing throughput is deliberate policy decisions, made for contractual or risk management or aggregation saturation or marketing reasons. Examples are Internet service providers rate limiting and bandwidth throttling and assignment of IPs to groups so as to minimize rather than maximize the throughput available to every user, but maximize the number of users that can be supported on one backbone.

Also, often, chips are not available to implement the fastest speeeds. AMD for instance does not support the 32-bit HyperTransport interface on any CPU it shipped as of end 2009, and WiMax service providers in the US are typically supporting only up to 4mbps as of end of 2009.

Choosing service providers or interfaces based on theoretical maxima is unwise, especially for commercial needs. Large scale data centers for instance would be more concerned with price per port to support the interface, wattage and heat considerations, and total cost of the solution. Scalability of the interface (some protocols such as SCSI and Ethernet now operate many orders of magnitude faster than when originally deployed) is one major factor as it prevents costly shifts to technologies that are not backward-compatible, often involuntarily or by surprise when a vendor abandons support for a proprietary system.

Conventions

By convention, bus and network speeds are denoted either in bit/s (bits per second) or byte/s (bytes per second). In general, parallel interfaces are quoted in byte/s and serial in bit/s. The more commonly used is shown below in bold type.

On devices like modems, bytes may be more than 8 bits long because they may be individually padded out with additional start and stop bits; the figures below will reflect this. Where channels use line codes (such as Ethernet, Serial ATA and PCI Express), quoted speeds are for the decoded signal.

The figures below are simplex speeds, which may conflict with the duplex speeds vendors sometimes use in promotional materials. Where two values are listed, the first value is the downstream rate and the second value is the upstream rate.

All quoted figures are in metric decimal units, where:

• 1 byte = 8 bits
• 1 kbit = 1,000 bits
• 1 Mbit = 1,000,000 bits
• 1 Gbit = 1,000,000,000 bits
• 1 kB = 1,000 bytes
• 1 MB = 1,000,000 bytes
• 1 GB = 1,000,000,000 bytes
• 1 TB = 1,000,000,000,000 bytes

Bandwidths

The figures below are grouped by network or bus type, then sorted within each group from lowest to highest bandwidth; gray shading indicates a lack of known implementations.

TTY/Teletypewriter or Telecommunications Device for the Deaf (TDD)

Modems/broadband connections

All modems are assumed to be in serial operation with 1 start bit, 8 data bits, no parity, and 1 stop bit (2 stop bits for 110-baud modems). Therefore, a total of 10 bits (11 bits for 110-baud modems) are needed to transmit each 8-bit byte. The "bytes" column reflects the net data transfer rate after the protocol overhead has been removed.

Mobile telephone interfaces

Wide area networks

Local area networks

Wireless networks

802.11 networks in infrastucture mode are half-duplex; all stations share the medium. In access point (infrastucture) mode, all traffic has to pass through the AP (Access Point). Thus, two stations on the same AP which are communicating with each other must have each and every frame transmitted twice: from the sender to the access point, then from the access point to the receiver. This approximately halves the effective bandwidth. In ad hoc mode devices communicate directly (like with a crossover cable) rather than to the network(like through a hub), losing some security functionality.

Wireless personal area networks

Computer buses

Portable

Storage

Peripheral

MAC to PHY

PHY to XPDR

Memory Interconnect/RAM buses

Dual channel bandwidths are theoretical maxima and do not always reflect real world performance. In many cases, performance may be closer to single channel operation (half the bandwidth).

Digital audio

Digital video interconnects

Speeds given are from the video source (e.g. video card) to receiving device (e.g. monitor) only. Out of band and reverse signaling channels are not included.

See also

• Bitrate
• Comparison of wireless data standards
• Comparison of mobile phone standards
• List of Internet access technology bit rates in the Digital bandwidth article
• OFDM system comparison table
• Spectral efficiency comparison table
• Sneakernet

Notes

External links

• Interconnection Speeds Compared
• Need for Speed: Theoretical Bandwidth Comparison — Contains a graph illustrating bandwidth speeds