Overview of 100G QSFP28 Optical Transceivers

by http://www.fiber-mart.com

QSFP28 fiber optic module has become the dominant form factor for 100G high-speed networks. The interconnect offers multiple channels of high-speed differential signals with data rates ranging from 25Gbps up to potentially 40Gbps, and meets 100Gbps Ethernet (4×25Gbps) and 100Gbps 4X InfiniBand Enhanced Data Rate (EDR) requirements. TARLUZ 100G QSFP28 optical transceiver including SR4, LR4, PSM4, CWDM4 and AOCs, complied with IEEE 802.3bm and SFF-8636, compatible with network device from different vendors, designed for applications of 100G Data Center Internal Network, Data Center Interconnection and Metro Network.
The following list is QSFP28 fiber optic transceivers form TARLUZ, it is able to compatible with the main network device provider like Cisco, HPE, Huawei, etc.
QSFP28 SR4: The QSFP28-SR4 optical module supports links of 70m over OM3 MMF and 100m over OM4 MMF with MPO-12 or MTP-12 connectors. This transceiver is a parallel 100G QSFP28 optical module with 4 independent transmit and receive channels each capable of 25Gb/s operation. The 100G QSFP28-SR4 modules are ideal for rack to rack connections in the datacenter and short reach telecom applications.The QSFP28-100G-eSR4 is extended version of QSFP for transmit over 300m.
QSFP28 PSM4: The 100G PSM4 specification defines requirements for a point-to-point 100 Gbps link over eight single mode fibers (4 transmit and 4 receive) of at least 500m, each transmitting at 25Gbps. Four identical and independent lanes are used for each signal direction. PSM4 does not need a MUX/DEMUX for each laser but it does need either a directly modulated DFB laser (DML) or an external modulator for each fiber. With an MTP interface, PSM4 modules can bus 100Gbps point-to-point over 2km or can be broken out into dual 50Gbps or quad 25Gbps links for linking to servers, storage and other subsystems.
QSFP28 CWDM4: The CWDM4 module uses Mux/Demux technologies with 4 lanes of 25 Gbps optically multiplexed onto and demultiplexed from duplex single-mode fiber. It is centered around the 1310nm band with 20nm channel spacing as defined by the ITU standard. With a reach of 2km, QSFP28 CWDM4 transmits 100G optical signals via a duplex LC interface.
QSFP28 LR4: This module is for longer span 100GbE deployment, such as connectivity between two buildings, QSFP28-LR4 with duplex LC fiber interface and transmitted over single-mode fiber cable. This LR4 module uses WDM technologies to achieve 100G transmission over four different wavelengths around 1310nm. It can support distances up to 10km.
QSFP28 ER4 Lite: QSFP28-ER4 Lite is a 100Gbps transceiver designed for optical communication applications compliant to Ethernet 100GBASE-ER4 Lite standard. The high performance cooled LAN WDM EA-DFB transmitters and high sensitivity APD receivers provide superior performance for 100Gigabit Ethernet applications up to 25km links without FEC and 32km links with FEC.

What are polarization maintaining fibers?

In the most common an optical fiber in which the polarization of linearly polarized light waves launched into the fiber is maintained during propagation, with little or no cross-coupling of optical power between the polarization modes.

 

In the most common an optical fiber in which the polarization of linearly polarized light waves launched into the fiber is maintained during propagation, with little or no cross-coupling of optical power between the polarization modes. Such fiber is used in special applications where preserving polarization is essential.

 

 What is polarization maintaining(PM) fibers ? 

 

Polarization Maintaining (PM) optical fiber is a key component of Fiber Optic Gyroscopes, devices that measure rotation in missiles, aircraft, ships and satellites. They are a type of interferometric sensor in which the phase difference between two light paths is measured.The polarization of light propagating in the fiber gradually changes in an uncontrolled (and wavelength-dependent) way, which also depends on any bending of the fiber and on its temperature. Specialised fibers are required to achieve optical performances, which are affected by the polarization of the light travelling through the fiber.optical fibers always exhibit some degree of birefringence, even if they have a circularly symmetric design, because in practice there is always some amount of mechanical stress or other effect which breaks the symmetry. As a consequence, the polarization of light propagating in the fiber gradually changes in an uncontrolled (and wavelength-dependent) way, which also depends on any bending of the fiber and on its temperature.

 

 Principle of polarization maintaining(PM) fibers

 

The mentioned problem can be fixed by using a polarization-maintaining fiber, which is not a fiber without birefringence, but on the contrary a specialty fiber with a strong built-in birefringence (high-birefringence fiber or HIBI fiber, PM fiber). In general, optical fiber telecommunications applications, PM fiber is used to guide light in a linearly polarised state from one place to another. To achieve this result, several conditions must be met. Input light must be highly polarised to avoid launching both slow and fast axis modes, a condition in which the output polarization state is unpredictable.

 

Provided that the polarization of light launched into the fiber is aligned with one of the birefringent axes, this polarization state will be preserved even if the fiber is bent. The physical principle behind this can be understood in terms of coherent mode coupling. The propagation constants of the two polarization modes are different due to the strong birefringence, so that the relative phase of such copropagating modes rapidly drifts away. Therefore, any disturbance along the fiber can effectively couple both modes only if it has a significant spatial Fourier component with a wavenumber which matches the difference of the propagation constants of the two polarization modes. If this difference is large enough, the usual disturbances in the fiber are too slowly varying to do effective mode coupling.In addition, connectors must have been installed on the PM fibers in such a way that internal stresses do not cause the electric field to be projected onto the unintended axis of the fiber.

 

Applications

 

PM optical fibers are used in special applications, such as in fiber optic sensing, interferometry and quantum key distribution. They are also commonly used in telecommunications for the connection between a source laser and a modulator, since the modulator requires polarized light as input. They are rarely used for long-distance transmission, because PM fiber is expensive and has higher attenuation than singlemode fiber.Optical fibers may be applied in measurements of electrical current, particularly as so-called optical current transformers. Electric current sensors, in which optical fibers are used are small, light, cheap and safe. Their sensitivity is, however, due to the restricted magnetootpic properties of optical fibers, rather small. Moreover, these sensors are susceptible to deformations of the optical fiber. An increase of their sensitivity consists in lengthening the distance of optical fiber on which the magnetic field acts.PM fibers are applied in devices where the polarization state cannot be allowed to drift, e.g. as a result of temperature changes. Examples are fiber interferometers and certain fiber lasers. A disadvantage of using such fibers is that usually an exact alignment of the polarization direction is required, which makes production more cumbersome. Also, propagation losses are higher than for standard fiber, and not all kinds of fibers are easily obtained in polarization-preserving form.

 

Fiber-MART offer polarization components can be utilized in high power optical amplifiers and optical transmission system, test and measurement. the launch conditions at the optical fiber end face must be consistent with the direction of the transverse major axis of the fiber cross section. Fiber-Mart Polarizing Beam Combiner/Splitter (PBC/PBS) is a compact high performance light wave component that combines two orthogonal polarization signals into one output fiber, and also can split the incoming light into two orthogonal states. We also supply the Isolator type (IPBC/IPBS) which provides both polarization beam combining and optical isolation in one integrated component.for more information,you can visit www.fiber-mart.com.pls feel free to contact with us for any question . E-mail: service@fiber-mart.com

OM4 fiber optic cabling

by http://www.fiber-mart.com

OM4 Specifications
How does OM4 compare to OM1, OM2, OM3, and single mode? There are significant differences between most of the standardized types of glass. A select few of the major attributes of these different glass types are shown below to highlight the differences.
It is important to note that OM4 glass is not necessarily designed to be a replacement for OM3. Despite the relatively long-standing availability of OM4, there are no plans to obsolete OM3 fiber optic cabling. For most systems, OM3 glass is sufficient to cover the bandwidth needs at the distances of the current installation base. Most system requirements can still be reliably and cost effectively achieved with OM3, and this glass type will remain available for the foreseeable future.
Despite the availability of OM4 glass, OM3 is quite capable of 40 and 100 Gb/s applications albeit at significantly shorter distances than OM4. The primary benefit that OM4 provides is additional reach at extended bandwidth at an overall cost still less than that of an OS2 singlemode system. In other words, OM4 provides a solution that allows more installations to avoid the significantly higher costs of singlemode systems.
OM4 Compatibility
Additionally, OM4 provides an opportunity to future-proof cabling infrastructure. OM4 is completely backwards-compatible with existing OM3 systems. As a result, these two grades of glass are interchangeable within the transmission distance limitations outlined above. The additional bandwidth and lower attenuation of OM4 provide additional insertion loss margin. As a result, users of OM4 gain additional safety margin to help compensate for less-than-ideal cabling installations as well as provide margin for degradation due to moves, adds, and changes over the life of the installation.
As increased bandwidth requirements are called out in new installations, particularly 40 and 100 Gb/s standards, transmission distances over fiber optic cables contained in existing infrastruOM4-LC-LC-Patch-Cord-1cture may become increasingly limited. Increasingly, these higher bandwidth system requirements have dictated a need to transition from cost-effective multi-mode systems to more costly single-mode systems. Until OM4 was formally specified, many next-generation 40 and 100 Gb/s applications would have had to make the leap to single-mode system solutions. OM4 effectively provides an additional layer of performance that supports these applications at longer distances, thereby limiting the number of installations that truly require OS2 singlemode fiber. OM4 can provide a minimum reach of 125m over multimode fiber within the 40 and 100 GbE standards.

When should i use OM5 Fiber

by http://www.fiber-mart.com

The need to sort through these permutations may partially explain the reportedly low number of OM5 deployments so far. Even cabling suppliers with OM5 in their portfolios note that most 40 and 100 Gigabit Ethernet links are likely to fall within the reach of OM4, making the extended reach of OM5 unnecessary.
For these reasons and others, some cabling suppliers have opted not to add OM5 to their lines. In a blog posted this past April, Gary Bernstein, senior director of product management for fiber and data center solutions at Leviton, described why his company doesn’t support OM5, stating:
The reach advantage of OM5 over OM4 is minimal.
OM5 won’t reduce costs. (OM5 fiber carries a cost premium, and 100-Gbps optics prices are in decline, reasons Bernstein).
It won’t enable higher port densities, since you can’t break out SWDM transmissions into their component parts the way you can with parallel fiber approaches.
A lot of large-scale data centers with a need for 40 or 100 Gigabit Ethernet have or will soon move to single-mode fiber anyway.
This is not to say that the fiber does not have its proponents, particularly for applications that require that extra bit of reach (see, for example, this whitepaper from CommScope). Meanwhile, there is an advantage beyond reach to OM5 and SWDM that could prove useful in future high-speed networks – the ability of one fiber to offer the transmission capacity that currently requires four in conventional use. At 40 or 100 Gbps, that ability could prove helpful when operating in space-constrained environments.
The need to sort through these permutations may partially explain the reportedly low number of OM5 deployments so far. Even cabling suppliers with OM5 in their portfolios note that most 40 and 100 Gigabit Ethernet links are likely to fall within the reach of OM4, making the extended reach of OM5 unnecessary.
For these reasons and others, some cabling suppliers have opted not to add OM5 to their lines. In a blog posted this past April, Gary Bernstein, senior director of product management for fiber and data center solutions at Leviton, described why his company doesn’t support OM5, stating:
The reach advantage of OM5 over OM4 is minimal.
OM5 won’t reduce costs. (OM5 fiber carries a cost premium, and 100-Gbps optics prices are in decline, reasons Bernstein).
It won’t enable higher port densities, since you can’t break out SWDM transmissions into their component parts the way you can with parallel fiber approaches.
A lot of large-scale data centers with a need for 40 or 100 Gigabit Ethernet have or will soon move to single-mode fiber anyway.
This is not to say that the fiber does not have its proponents, particularly for applications that require that extra bit of reach (see, for example, this whitepaper from CommScope). Meanwhile, there is an advantage beyond reach to OM5 and SWDM that could prove useful in future high-speed networks – the ability of one fiber to offer the transmission capacity that currently requires four in conventional use. At 40 or 100 Gbps, that ability could prove helpful when operating in space-constrained environments.

Fujikura 22S cladding alignment fusion splicer

by http://www.fiber-mart.com

AFL has introduced the Fujikura 22S active cladding alignment fusion splicer. With this model’s moveable v-grooves, splicer errors due to dust and other contaminants are virtually eliminated, says the company. Removable sheath clamps allow the use of fiber holders, and the unit’s large monitor provides a crystal clear image, even in bright sunlight.
“Fujikura continues to improve upon fusion splicing technology by incorporating newer features that make splicing easier,” comments Greg Pickeral, product manager for AFL’s fusion splicing systems. “The Fujikura 22S incorporates many of the advanced features of our more expensive models yet retains the quality and reliability they are known for.”
The fully ruggedized Fujikura 22S chassis provides for shock, dust and moisture protection, while the model’s two camera observation system provides for accurate fiber alignment and loss estimation calculations. Additional features include a long-life battery that provides power for up to 200 splice cycles (including the application of the splice sleeve), and an electrode life which has been extended to 5,000 splices, minimizing downtime for replacement and stabilizations. The unit’s transit case and work tray provide multiple options for utilizing workspace.
Ideal for field splicing, the 22S maintains high quality in the most extreme environments. Software updates are available via the Internet allowing users to quickly update their software as new splice programs become available. The Fujikura 22S is also fully compatible with the company’s FUSEConnect line of fusion installable connectors.

Three connection modes of the switch

There are three main ways to connect switches: cascading, stacking, and clustering. Cascade mode is simple to implement, just an ordinary twisted pair can be, cost savings and basically not limited by the distance. The investment in the stacking method is relatively large and can only be connected within a short distance, which is difficult to achieve. Cluster connection means that multiple interconnected (cascaded or stacked) switches are managed as a logical device.

There are three main ways to connect switches: cascading, stacking, and clustering. Cascade mode is simple to implement, just an ordinary twisted pair can be, cost savings and basically not limited by the distance. The investment in the stacking method is relatively large and can only be connected within a short distance, which is difficult to achieve. Cluster connection means that multiple interconnected (cascaded or stacked) switches are managed as a logical device.

tr
The stacking mode has better performance than the cascaded mode, and the signal is not easily depleted. Through the stacking mode, multiple switches can be managed in a centralized manner, which greatly reduces the management workload. If you really need to use cascading, you can also use the Uplink port. Connection method. Because this can guarantee the signal intensity to the greatest extent, if it is the connection between ordinary ports, it will certainly make the network signal seriously damaged.

1. Switch cascading
This is the most common way to connect multiple switches. It connects through the UpLink on the switch. It should be noted that the switches cannot be cascaded without limitation. Cascading over a certain number of switches will eventually cause broadcast storms, which will lead to a serious drop in network performance. Cascading is further divided into using ordinary port cascading and using Uplink port cascading.

 

2

1.1. Use ordinary port cascading

The so-called ordinary port is through a switch of a common port (such as RJ-45 port) to connect.
In the past, it was necessary to use the reverse connection. Now the two ends of the network cable are 568b line sequence. The jumper line is 568a line and 568b line sequence. According to the need, the old version of the device will distinguish the direct line from the crossover line. Now the devices are all common. What kind of consequences, the switch can automatically identify, and only the line can be wrong.

 

1.2 Use Uplink port cascading

In all switch ports, an Uplink port is included next to it. This port is provided exclusively for upstream connections. Simply connect the port to a port on the other switch except for the “Uplink port” through a straight-through twisted pair (note that it is not the Uplink port that is connected to each other).

33

2. Switch stack
This type of connection is mainly used in large networks where port requirements are relatively large. The stacking of switches is the quickest and most convenient way to expand ports. At the same time, the bandwidth after stacking is several-tenths of the speed of a single switch port. But not all switches support stacking, depending on whether the switch’s brand or model supports stacking. It is mainly connected through a dedicated connection cable provided by the manufacturer from the “UP” stack port of one switch to the “DOWN” stack port of another switch. All switches in a stack can be managed as a single switch.

Stacked switches are limited by the type and mutual distance. First, the stack switches must support stacking; in addition, the stacked connection cables provided by the manufacturers are generally around 1M, so the stacking function can only be used within a short distance.

34

3. Cluster
In the so-called cluster, multiple interconnected (cascaded or stacked) switches are managed as a logical device. In a cluster, there is generally only one switch that functions as a management switch,which is called a command switch. It can manage several other switches. In the network, these switches only need to occupy one IP address (only required by the command switch). Under unified management of the command switch, multiple switches in the cluster work together to greatly reduce management intensity.

It should be noted that different manufacturers have different implementations for clusters, and generally manufacturers use proprietary protocols to implement clusters. This determines the cluster technology has its limitations. Switches of different manufacturers can be cascaded but cannot be clustered.

444
Switching, stacking, and clustering are three different technologies. Cascading and stacking are prerequisites for implementing clusters. Clusters are used for cascading and stacking; cascading and stacking are implemented based on hardware; clusters are implemented based on software; cascading and stacking are sometimes similar (especially cascading and virtual Stacking), sometimes very different (cascade and real stacking).Please feel free to contact Fiber-Mart if you have any needs or questions.we will provide you with the most professional service.