Small Form-factor Pluggable (SFP) is a compact, hot-pluggable network interface module format used for both telecommunication and data communications applications. An SFP interface on networking hardware is a modular slot for a media-specific transceiver, such as for a fiber-optic cable or a copper cable. [1] The advantage of using SFPs compared to fixed interfaces (e.g. modular connectors in Ethernet switches) is that individual ports can be equipped with different types of transceivers as required, with the majority including optical line terminals, network cards, switches and routers.
The form factor and electrical interface are specified by a multi-source agreement (MSA) under the auspices of the Small Form Factor Committee. [2] The SFP replaced the larger gigabit interface converter (GBIC) in most applications, and has been referred to as a Mini-GBIC by some vendors. [3]
SFP transceivers exist supporting synchronous optical networking (SONET), Gigabit Ethernet, Fibre Channel, PON, and other communications standards. At introduction, typical speeds were 1 Gbit/s for Ethernet SFPs and up to 4 Gbit/s for Fibre Channel SFP modules. [4] In 2006, SFP+ specification brought speeds up to 10 Gbit/s and the later SFP28 iteration, introduced in 2014, [5] is designed for speeds of 25 Gbit/s. [6]
A slightly larger sibling is the four-lane Quad Small Form-factor Pluggable (QSFP). The additional lanes allow for speeds 4 times their corresponding SFP. In 2014, the QSFP28 variant was published allowing speeds up to 100 Gbit/s. [7] In 2019, the closely related QSFP56 was standardized [8] doubling the top speeds to 200 Gbit/s with products already selling from major vendors. [9] There are inexpensive adapters allowing SFP transceivers to be placed in a QSFP port.
Both a SFP-DD, [10] which allows for 100 Gbit/s over two lanes, as well as a QSFP-DD [11] specifications, which allows for 400 Gbit/s over eight lanes, have been published. [12] These use a form factor which is directly backward compatible to their respective predecessors. [13]
An even larger sibling, the OSFP (Octal Small Format Pluggable) has products being released in 2022 [14] capable of 800 Gbit/s links between network equipment. It is a slightly larger version than the QSFP form factor allowing for larger power outputs. The OSFP standard was initially announced in 2016 [15] with the 4.0 version released in 2021 allowing for 800 Gbit/s via 8×100 Gbit/s electrical data lanes. [16] Its proponents say a low-cost adapter will allow for backwards compatibility with QSFP modules. [17]
SFP transceivers are available with a variety of transmitter and receiver specifications, allowing users to select the appropriate transceiver for each link to provide the required optical or electrical reach over the available media type (e.g. twisted pair or twinaxial copper cables, multi-mode or single-mode fiber cables). Transceivers are also designated by their transmission speed. SFP modules are commonly available in several different categories.
Name | Nominal speed | Lanes | Standard | Introduced | Backward-compatible | PHY interface | Connector |
---|---|---|---|---|---|---|---|
SFP | 100 Mbit/s | 1 | SFF INF-8074i | 2001-05-01 | None | MII | LC, RJ45 |
SFP | 1 Gbit/s | 1 | SFF INF-8074i | 2001-05-01 | 100 Mbit/s SFP* | SGMII | LC, RJ45 |
cSFP | 1 Gbit/s | 2 | LC | ||||
SFP+ | 10 Gbit/s | 1 | SFF SFF-8431 4.1 | 2009-07-06 | SFP | XGMII | LC, RJ45 |
SFP28 | 25 Gbit/s | 1 | SFF SFF-8402 | 2014-09-13 | SFP, SFP+ | LC | |
SFP56 | 50 Gbit/s | 1 | SFP, SFP+, SFP28 | LC | |||
SFP-DD | 100 Gbit/s | 2 | SFP-DD MSA [18] | 2018-01-26 | SFP, SFP+, SFP28, SFP56 | LC | |
SFP112 | 100 Gbit/s | 1 | 2018-01-26 | SFP, SFP+, SFP28, SFP56 | LC | ||
SFP-DD112 | 200 Gbit/s | 2 | 2018-01-26 | SFP, SFP+, SFP28, SFP56, SFP-DD, SFP112 | LC | ||
QSFP types | |||||||
QSFP | 4 Gbit/s | 4 | SFF INF-8438 | 2006-11-01 | None | GMII | |
QSFP+ | 40 Gbit/s | 4 | SFF SFF-8436 | 2012-04-01 | None | XGMII | LC, MTP/MPO |
QSFP28 | 50 Gbit/s | 2 | SFF SFF-8665 | 2014-09-13 | QSFP+ | LC | |
QSFP28 | 100 Gbit/s | 4 | SFF SFF-8665 | 2014-09-13 | QSFP+ | LC, MTP/MPO-12 | |
QSFP56 | 200 Gbit/s | 4 | SFF SFF-8665 | 2015-06-29 | QSFP+, QSFP28 | LC, MTP/MPO-12 | |
QSFP112 | 400 Gbit/s | 4 | SFF SFF-8665 | 2015-06-29 | QSFP+, QSFP28, QSFP56 | LC, MTP/MPO-12 | |
QSFP-DD | 400 Gbit/s | 8 | SFF INF-8628 | 2016-06-27 | QSFP+, QSFP28, [19] QSFP56 | LC, MTP/MPO-16 |
Note that the QSFP/QSFP+/QSFP28/QSFP56 are designed to be electrically backward compatible with SFP/SFP+/SFP28 or SFP56 respectively. Using a simple adapter or a special direct attached cable it is possible to connect those interfaces together using just one lane instead of four provided by the QSFP/QSFP+/QSFP28/QSFP56 form factor. The same applies to the QSFP-DD form factor with 8 lanes which can work downgraded to 4/2/1 lanes.
The SFP+ (enhanced small form-factor pluggable) is an enhanced version of the SFP that supports data rates up to 16 Gbit/s. The SFP+ specification was first published on May 9, 2006, and version 4.1 was published on July 6, 2009. [31] SFP+ supports 8 Gbit/s Fibre Channel, 10 Gigabit Ethernet and Optical Transport Network standard OTU2. It is a popular industry format supported by many network component vendors. Although the SFP+ standard does not include mention of 16 Gbit/s Fibre Channel, it can be used at this speed. [32] Besides the data rate, the major difference between 8 and 16 Gbit/s Fibre Channel is the encoding method. The 64b/66b encoding used for 16 Gbit/s is a more efficient encoding mechanism than 8b/10b used for 8 Gbit/s, and allows for the data rate to double without doubling the line rate. 16GFC doesn't really use 16 Gbit/s signaling anywhere. It uses a 14.025 Gbit/s line rate to achieve twice the throughput of 8GFC. [33]
SFP+ also introduces direct attach for connecting two SFP+ ports without dedicated transceivers. Direct attach cables (DAC) exist in passive (up to 7 m), active (up to 15 m), and active optical (AOC, up to 100 m) variants.
10 Gbit/s SFP+ modules are exactly the same dimensions as regular SFPs, allowing the equipment manufacturer to re-use existing physical designs for 24 and 48-port switches and modular line cards. In comparison to earlier XENPAK or XFP modules, SFP+ modules leave more circuitry to be implemented on the host board instead of inside the module. [34] Through the use of an active electronic adapter, SFP+ modules may be used in older equipment with XENPAK ports [35] and X2 ports. [36] [37]
SFP+ modules can be described as limiting or linear types; this describes the functionality of the inbuilt electronics. Limiting SFP+ modules include a signal amplifier to re-shape the (degraded) received signal whereas linear ones do not. Linear modules are mainly used with the low bandwidth standards such as 10GBASE-LRM; otherwise, limiting modules are preferred. [38]
SFP28 is a 25 Gbit/s interface which evolved from the 100 Gigabit Ethernet interface which is typically implemented with 4 by 25 Gbit/s data lanes. Identical in mechanical dimensions to SFP and SFP+, SFP28 implements one 28 Gbit/s lane [39] accommodating 25 Gbit/s of data with encoding overhead. [40]
SFP28 modules exist supporting single- [41] or multi-mode [42] fiber connections, active optical cable [43] and direct attach copper. [44] [45]
The compact small form-factor pluggable (cSFP) is a version of SFP with the same mechanical form factor allowing two independent bidirectional channels per port. It is used primarily to increase port density and decrease fiber usage per port. [46] [47]
The small form-factor pluggable double density (SFP-DD) multi-source agreement is a standard published in 2019 for doubling port density. According to the SFD-DD MSA website: "Network equipment based on the SFP-DD will support legacy SFP modules and cables, and new double density products." [48] SFP-DD uses two lanes to transmit.
Currently, the following speeds are defined:
Quad Small Form-factor Pluggable (QSFP) transceivers are available with a variety of transmitter and receiver types, allowing users to select the appropriate transceiver for each link to provide the required optical reach over multi-mode or single-mode fiber.
Switch and router manufacturers implementing QSFP+ ports in their products frequently allow for the use of a single QSFP+ port as four independent 10 Gigabit Ethernet connections, greatly increasing port density. For example, a typical 24-port QSFP+ 1U switch would be able to service 96x10GbE connections. [58] [59] [60] There also exist fanout cables to adapt a single QSFP28 port to four independent 25 Gigabit Ethernet SFP28 ports (QSFP28-to-4×SFP28) [61] as well as cables to adapt a single QSFP56 port to four independent 50 Gigabit Ethernet SFP56 ports (QSFP56-to-4×SFP56). [62]
SFP sockets are found in Ethernet switches, routers, firewalls and network interface cards. They are used in Fibre Channel host adapters and storage equipment. Because of their low cost, low profile, and ability to provide a connection to different types of optical fiber, SFP provides such equipment with enhanced flexibility.
SFP sockets and transceivers are also used for long-distance serial digital interface (SDI) transmission. [63]
The SFP transceiver is not standardized by any official standards body, but rather is specified by a multi-source agreement (MSA) among competing manufacturers. The SFP was designed after the GBIC interface, and allows greater port density (number of transceivers per given area) than the GBIC, which is why SFP is also known as mini-GBIC.
However, as a practical matter, some networking equipment manufacturers engage in vendor lock-in practices whereby they deliberately break compatibility with generic SFPs by adding a check in the device's firmware that will enable only the vendor's own modules. [64] Third-party SFP manufacturers have introduced SFPs with EEPROMs which may be programmed to match any vendor ID. [65]
Color | Standard | Media | Wavelength | Notes |
---|---|---|---|---|
Black | INF-8074 | Multimode | 850 nm | |
Beige | INF-8074 | Multimode | 850 nm | |
Black | INF-8074 | Multimode | 1300 nm | |
Blue | INF-8074 | Singlemode | 1310 nm | |
Red | proprietary (non SFF) | Singlemode | 1310 nm | Used on 25GBASE-ER [66] |
Green | proprietary (non SFF) | Singlemode | 1550 nm | Used on 100BASE-ZE |
Red | proprietary (non SFF) | Singlemode | 1550 nm | Used on 10GBASE-ER |
White | proprietary (non SFF) | Singlemode | 1550 nm | Used on 10GBASE-ZR |
Color | Standard | Wavelength | Notes |
---|---|---|---|
Grey | 1270 nm | ||
Grey | 1290 nm | ||
Grey | 1310 nm | ||
Violet | 1330 nm | ||
Blue | 1350 nm | ||
Green | 1370 nm | ||
Yellow | 1390 nm | ||
Orange | 1410 nm | ||
Red | 1430 nm | ||
Brown | 1450 nm | ||
Grey | 1470 nm | ||
Violet | 1490 nm | ||
Blue | 1510 nm | ||
Green | 1530 nm | ||
Yellow | 1550 nm | ||
Orange | 1570 nm | ||
Red | 1590 nm | ||
Brown | 1610 nm |
Name | Standard | Side A Color TX | Side A wavelength TX | Side B Color TX | Side B wavelength TX | Notes |
---|---|---|---|---|---|---|
1000BASE-BX | Blue | 1310 nm | Purple | 1490 nm | ||
1000BASE-BX | Blue | 1310 nm | Yellow | 1550 nm | ||
10GBASE-BX 25GBASE-BX | Blue | 1270 nm | Red | 1330 nm | ||
10GBASE-BX | White | 1490 nm | White | 1550 nm |
Color | Standard | Wavelength | Multiplexing | Notes |
---|---|---|---|---|
Beige | INF-8438 | 850 nm | No | |
Blue | INF-8438 | 1310 nm | No | |
White | INF-8438 | 1550 nm | No |
SFP transceivers are right-handed: From their perspective, they transmit on the right and receive on the left. When looking into the optical connectors, transmission comes from the left and reception is on the right. [68]
The SFP transceiver contains a printed circuit board with an edge connector with 20 pads that mate on the rear with the SFP electrical connector in the host system. The QSFP has 38 pads including 4 high-speed transmit data pairs and 4 high-speed receive data pairs. [52] [53]
Pad | Name | Function |
---|---|---|
1 | VeeT | Transmitter ground |
2 | Tx_Fault | Transmitter fault indication |
3 | Tx_Disable | Optical output disabled when high |
4 | SDA | 2-wire serial interface data line (using the CMOS EEPROM protocol defined for the ATMEL AT24C01A/02/04 family [69] ) |
5 | SCL | 2-wire serial interface clock |
6 | Mod_ABS | Module absent, connection to VeeT or VeeR in the module indicates module presence to host |
7 | RS0 | Rate select 0 |
8 | Rx_LOS | Receiver loss of signal indication |
9 | RS1 | Rate select 1 |
10 | VeeR | Receiver ground |
11 | VeeR | Receiver ground |
12 | RD- | Inverted received data |
13 | RD+ | Received data |
14 | VeeR | Receiver ground |
15 | VccR | Receiver power (3.3 V, max. 300 mA) |
16 | VccT | Transmitter power (3.3 V, max. 300 mA) |
17 | VeeT | Transmitter ground |
18 | TD+ | Transmit data |
19 | TD- | Inverted transmit data |
20 | VeeT | Transmitter ground |
Pad | Name | Function |
---|---|---|
1 | GND | Ground |
2 | Tx2n | Transmitter inverted data input |
3 | Tx2p | Transmitter non-inverted data input |
4 | GND | Ground |
5 | Tx4n | Transmitter inverted data input |
6 | Tx4p | Transmitter non-inverted data input |
7 | GND | Ground |
8 | ModSelL | Module select |
9 | ResetL | Module reset |
10 | Vcc-Rx | +3.3 V receiver power supply |
11 | SCL | Two-wire serial interface clock |
12 | SDA | Two-wire serial interface data |
13 | GND | Ground |
14 | Rx3p | Receiver non-inverted data output |
15 | Rx3n | Receiver inverted data output |
16 | GND | Ground |
17 | Rx1p | Receiver non-inverted data output |
18 | Rx1n | Receiver inverted data output |
19 | GND | Ground |
20 | GND | Ground |
21 | Rx2n | Receiver inverted data output |
22 | Rx2p | Receiver non-inverted data output |
23 | GND | Ground |
24 | Rx4n | Receiver inverted data output |
25 | Rx4p | Receiver non-inverted data output |
26 | GND | Ground |
27 | ModPrsL | Module present |
28 | IntL | Interrupt |
29 | Vcc-Tx | +3.3 V transmitter power supply |
30 | Vcc1 | +3.3 V power supply |
31 | LPMode | Low power mode |
32 | GND | Ground |
33 | Tx3p | Transmitter non-inverted data input |
34 | Tx3n | Transmitter inverted data input |
35 | GND | Ground |
36 | Tx1p | Transmitter non-inverted data input |
37 | Tx1n | Transmitter inverted data input |
38 | GND | Ground |
The physical dimensions of the SFP transceiver (and its subsequent faster variants) are narrower than the later QSFP counterparts, which allows for SFP transceivers to be placed in QSFP ports via an inexpensive adapter. Both are smaller than the XFP transceiver.
SFP [2] | QSFP [52] | XFP [70] | ||||
---|---|---|---|---|---|---|
mm | in | mm | in | mm | in | |
Height | 8.5 | 0.33 | 8.5 | 0.33 | 8.5 | 0.33 |
Width | 13.4 | 0.53 | 18.35 | 0.722 | 18.35 | 0.722 |
Depth | 56.5 | 2.22 | 72.4 | 2.85 | 78.0 | 3.07 |
The SFP MSA defines a 256-byte memory map into an EEPROM describing the transceiver's capabilities, standard interfaces, manufacturer, and other information, which is accessible over a serial I²C interface at the 8-bit address 0b1010000X (0xA0). [71]
Modern optical SFP transceivers support standard digital diagnostics monitoring (DDM) functions. [72] This feature is also known as digital optical monitoring (DOM). This capability allows monitoring of the SFP operating parameters in real time. Parameters include optical output power, optical input power, temperature, laser bias current, and transceiver supply voltage. In network equipment, this information is typically made available via Simple Network Management Protocol (SNMP). A DDM interface allows end users to display diagnostics data and alarms for optical fiber transceivers and can be used to diagnose why a transceiver is not working.
A network switch is networking hardware that connects devices on a computer network by using packet switching to receive and forward data to the destination device.
In computer networking, Fast Ethernet physical layers carry traffic at the nominal rate of 100 Mbit/s. The prior Ethernet speed was 10 Mbit/s. Of the Fast Ethernet physical layers, 100BASE-TX is by far the most common.
In computer networking, Gigabit Ethernet is the term applied to transmitting Ethernet frames at a rate of a gigabit per second. The most popular variant, 1000BASE-T, is defined by the IEEE 802.3ab standard. It came into use in 1999, and has replaced Fast Ethernet in wired local networks due to its considerable speed improvement over Fast Ethernet, as well as its use of cables and equipment that are widely available, economical, and similar to previous standards. The first standard for faster 10 Gigabit Ethernet was approved in 2002.
Fibre Channel (FC) is a high-speed data transfer protocol providing in-order, lossless delivery of raw block data. Fibre Channel is primarily used to connect computer data storage to servers in storage area networks (SAN) in commercial data centers.
Gigabit interface converter (GBIC) is a standard for transceivers. First defined in 1995, it was used with Gigabit Ethernet and Fibre Channel. By standardizing on a hot swappable electrical interface, a single gigabit port can support a wide range of physical media, from copper to long-wave single-mode optical fiber, at lengths of hundreds of kilometers.
XENPAK is a multisource agreement (MSA), instigated by Agilent Technologies and Agere Systems, that defines a fiber-optic or wired transceiver module which conforms to the 10 Gigabit Ethernet (10GbE) standard of the Institute of Electrical and Electronics Engineers (IEEE) 802.3 working group. The MSA group received input from both transceiver and equipment manufacturers during the definition process. XENPAK has been replaced by more compact devices providing the same functionality.
Multi-mode optical fiber is a type of optical fiber mostly used for communication over short distances, such as within a building or on a campus. Multi-mode links can be used for data rates up to 800 Gbit/s. Multi-mode fiber has a fairly large core diameter that enables multiple light modes to be propagated and limits the maximum length of a transmission link because of modal dispersion. The standard G.651.1 defines the most widely used forms of multi-mode optical fiber.
The physical-layer specifications of the Ethernet family of computer network standards are published by the Institute of Electrical and Electronics Engineers (IEEE), which defines the electrical or optical properties and the transfer speed of the physical connection between a device and the network or between network devices. It is complemented by the MAC layer and the logical link layer. An implementation of a specific physical layer is commonly referred to as PHY.
Twinaxial cabling, or twinax, is a type of cable similar to coaxial cable, but with two inner conductors in a twisted pair instead of one. Due to cost efficiency it is becoming common in modern (2013) very-short-range high-speed differential signaling applications.
40 Gigabit Ethernet (40GbE) and 100 Gigabit Ethernet (100GbE) are groups of computer networking technologies for transmitting Ethernet frames at rates of 40 and 100 gigabits per second (Gbit/s), respectively. These technologies offer significantly higher speeds than 10 Gigabit Ethernet. The technology was first defined by the IEEE 802.3ba-2010 standard and later by the 802.3bg-2011, 802.3bj-2014, 802.3bm-2015, and 802.3cd-2018 standards. The first succeeding Terabit Ethernet specifications were approved in 2017.
The current portfolio of PowerConnect switches are now being offered as part of the Dell Networking brand: information on this page is an overview of all current and past PowerConnect switches as per August 2013, but any updates on current portfolio will be detailed on the Dell Networking page.
The Cisco Nexus series switches are modular and fixed port network switches designed for the data center. Cisco Systems introduced the Nexus Series of switches on January 28, 2008. The first chassis in the Nexus 7000 family is a 10-slot chassis with two supervisor engine slots and eight I/O module slots at the front, as well as five crossbar switch fabric modules at the rear. Beside the Nexus 7000 there are also other models in the Nexus range.
10 Gigabit Ethernet is a group of computer networking technologies for transmitting Ethernet frames at a rate of 10 gigabits per second. It was first defined by the IEEE 802.3ae-2002 standard. Unlike previous Ethernet standards, 10GbE defines only full-duplex point-to-point links which are generally connected by network switches; shared-medium CSMA/CD operation has not been carried over from the previous generations of Ethernet standards so half-duplex operation and repeater hubs do not exist in 10GbE. The first standard for faster 100 Gigabit Ethernet links was approved in 2010.
The C form-factor pluggable is a multi-source agreement to produce a common form-factor for the transmission of high-speed digital signals. The c stands for the Latin letter C used to express the number 100 (centum), since the standard was primarily developed for 100 Gigabit Ethernet systems.
The XFP is a standard for transceivers for high-speed computer network and telecommunication links that use optical fiber. It was defined by an industry group in 2002, along with its interface to other electrical components, which is called XFI.
The Dell blade server products are built around their M1000e enclosure that can hold their server blades, an embedded EqualLogic iSCSI storage area network and I/O modules including Ethernet, Fibre Channel and InfiniBand switches.
Dell Networking is the name for the networking portfolio of Dell. In the first half of 2013, Dell started to rebrand their different existing networking product brands to Dell Networking. Dell Networking is the name for the networking equipment that was known as Dell PowerConnect, as well as the Force10 portfolio.
25 Gigabit Ethernet and 50 Gigabit Ethernet are standards for Ethernet connectivity in a datacenter environment, developed by IEEE 802.3 task forces 802.3by and 802.3cd and are available from multiple vendors.
An optical module is a typically hot-pluggable optical transceiver used in high-bandwidth data communications applications. Optical modules typically have an electrical interface on the side that connects to the inside of the system and an optical interface on the side that connects to the outside world through a fiber optic cable. The form factor and electrical interface are often specified by an interested group using a multi-source agreement (MSA). Optical modules can either plug into a front panel socket or an on-board socket. Sometimes the optical module is replaced by an electrical interface module that implements either an active or passive electrical connection to the outside world. A large industry supports the manufacturing and use of optical modules.
Coherent optical module refers to a typically hot-pluggable coherent optical transceiver that uses coherent modulation (BPSK/QPSK/QAM) rather than amplitude modulation (RZ/NRZ/PAM4) and is typically used in high-bandwidth data communications applications. Optical modules typically have an electrical interface on the side that connects to the inside of the system and an optical interface on the side that connects to the outside world through a fiber optic cable. The technical details of coherent optical modules were proprietary for many years, but have recently attracted efforts by multi-source agreement (MSA) groups and a standards development organizations such as the Optical Internetworking Forum. Coherent optical modules can either plug into a front panel socket or an on-board socket. Coherent optical modules form a smaller piece of a much larger optical module industry.
With the 800G spec completed, group is developing specification for 1600G modules