DWDM Solutions for Arista 7500E Series Switches

Nowadays, the deployment of DWDM solution has been hotly debated in many enterprise networks, especially in the new Lay2 and Lay3 equipment like Arista 7500E series switches. For many enterprises, DWDM network solutions are undoubtedly the best choices of action, because they can provide a scalable and elastic solution for the enterprise that offered high bandwidth and data separation. This article will demonstrate DWDM solutions to Arista 7500E switches which are the foundation of two-tier open networking solutions for cloud data centers.

Analysis of DWDM System

DWDM (Dense Wavelength Division Multiplexing) is a technology allowing high throughput capacity over longer distances commonly ranging between 44-88 channels and transferring data rates from 100 Mbps up to 100 Gbps per wavelength. For intra-datacenter solutions, an endpoint connection often uses multimode (850 nm) for short ranges and single mode (1310 nm) for longer ranges. The DWDM node converts this local connection to a channelized frequency or wavelength, which is then multiplexed with other wavelength and transmitted over a single fiber connection.

A key advantage of DWDM is that it’s bitrate independent. DWDM-based networks can transmit data in IP, ATM, SONET, SDH and Ethernet. Therefore, DWDM systems can carry different types of traffic at different speeds over an optical channel. Voice transmission, email, video and multimedia data are just some examples of services which can be simultaneously transmitted in DWDM systems.

Arista 7500E 100G DWDM Line Card

With full support for Layer2 and Layer3 protocols, Arista 7500E series switch is the ideal option for the network spine for two tier data centers applications. Arista 7500E especially provides the perfect resolution for high bandwidth Metro and long-haul DCI solutions with the 6-port DWDM line card. It has great advantage to migrate from existing 10G DWDM to 100G coherent line side modules. The 7500E series DWDM line card provides six 100G ports with coherent 100G tunable optics, which enables customers to connect directly into existing WDM MUX module without the need to add transceivers, which can save cost and space to a large extent. The coherent optics use C-band region wavelengths and offer a cost efficient solution for up to 96 channels of 100Gb over a single dark fiber pair.

Use Cases for Arista 7500E DWDM Card

• Less Than 80 km Dark Fiber Connection
  For distance less than 80 km, Arista 7500E switch with DWDM line cards can directly terminate a dark fiber connection with a pair of passive DWDM Mux, thus achieving a point-to-point connection between two locations.

• Between 80 km and 150 km Connection
  For distance greater than 80 km but less than 150 km, losses occurred during the process of transmission should be considered. In order to boost the power level, an EDFA (Erbium Doped Fiber Amplifier) is used to gain flatness, noise level, and output power, which is typically capable of gains of 30 dB or more and output power of +17 dB or more. With the use of EDFA, the signal can be boosted into a certain power level, thus achieving distances of up to 150 km.

Conclusion

The Arista 7500E series DWDM solution offers a cost-effective solution for transporting scalable and massive volumes of traffic, and enhances the 7500E system by providing high performance 100G DWDM port density with the same rich features and dedicated secure encryption in compact and power-efficient systems. Enterprises can easily migrate existing metro and long-haul DWDM networks to add new 100G capacities, thus expanding Layer2 and Layer3 services.

Originally published at http://www.china-cable-suppliers.com/dwdm-solutions-arista-7500e-series-switches.html

Using EDFA Amplifier for Long-Haul CATV Systems

With Laser technology combining with fiber optic technology, CATV systems in the field of optical communication have demonstrated unprecedented and irreplaceable achievements in the past few decades. When transmitting optical signals with fibers, fiber attenuation is the main factor that limits the transmission distance. EDFA (Erbium-Doped Fiber Amplifier) designed for CATV long-haul transmission avoids the conversion of optic-electric-optic in CATV long-haul transmission. It amplifies low signal power into high signal power, thus extending transmission distance. This post analyzes EDFA configurations and the utilization in long-haul CATV systems.

EDFA Leading Position in CATV Systems

EDFA is one of the most prominent achievements in fiber optic transmission technology over the past decade. Because it cleverly combined the laser technology and optical fiber manufacturing technology in the CATV systems and its applications were then rapidly expanded. Originally PDFA and EDFA amplifiers were equally used for CATV systems, but today, EDFA has completely replaced PDFA and become the primary device for fiber optic transmission systems. Why EDFA has leading position on CATV systems? Because EDFA noise and distortion characteristics are better, and its superior characteristics can be clearly seen in the following:

• Operates at wavelength of 1550nm, consistent with C-band where fiber has the lowest loss
• Has higher saturation output power, useful in systems requiring transmission up to 100 km or systems requiring the optical signal to be split to multiple fiber optic receivers
• The signal gain spectrum is wide up to 30nm or more, can be used for broadband signal amplification, especially for WDM (wavelength division multiplexing) system, ideal for radio and data services networks
• Has user friendly interface RS232, easy to control and monitor with computers
• Low noise figure with high stability

EDFA Configurations

The configuration of a co-propagating EDFA is shown in Figure 1. The optical pump is combined with the optical signal into the erbium-doped fiber with a wavelength division multiplexer. A second multiplexer removes residual pump light from the fiber. An in-line optical filter provides additional insurance that pump light does not reach the output of the optical amplifier. An optical isolator is used to prevent reflected light from other portions of the optical system from entering the amplifier.

An EDFA with a counter propagating pump is pictured in Figure 2. The copropagating geometry produces an amplifier with less noise and less output power. The counter propagating geometry produces a noisier amplifier with high output power. A compromise can be made by combining the co- and counter-propagating geometries in a bi-directional configuration.

A Typical CATV System Using EDFA

Figure 3 illustrates a basic long-haul CATV transmission system designed to carry 77 channels of CATV signals for 100 km in a basic point-to-point configuration.

As you can see in Figure 3, the local CATV provider sends 77 channels of CATV signals at the transmitting side. After processing and RF combining, those multiple signals are combined into one channel of CATV signal with the wavelengths of 1550 nm. It transmits over a single-mode optical fiber to 50 km. An EDFA amplifier is used at the middle point to amplify the signals to a certain power level, continuing to transmit over a single mode fiber to 100 km. At the receiving side, the 1550nm CATV channel is split into multiple channels of 1550nm CATV signals, serving multiple hotel cable TV users.

FS.COM CATV EDFA Optical Amplifiers List

EDFA has undoubtedly received wide interest for CATV applications because of its high output power, low distortion and low noise capability. FS.COM supplies optical amplifiers including CATV EDFA, SDH EDFA, DWDM amplifier, etc. The following table lists FS.COM CATV EDFA amplifiers which are available with range of output power from 13 dBm to 24 dBm to meet the requirements of a high-density solution for the large-scale distribution of broadband CATV video and data signals to video overlay receivers in a FTTH/FTTP or PON system.

Originally published at http://www.china-cable-suppliers.com/using-edfa-amplifier-long-haul-catv-systems.html

Evolution of Optical Wavelength Bands

As fiber optic networks have developed for higher speeds, longer distances, and wavelength-division multiplexing (WDM), fibers have been used in new wavelength ranges, namely "bands". Fiber transmission bands have been defined and standardized, from the original O-band to the U/XL-bands. This article will mainly illustrate the evolution of the typical fiber transmission bands used for different optical telecom systems.

Among these bands, the O-band, also called the Original-band, was the first band used in optical telecommunication because of the small pulse broadening (small dispersion); Single-mode fiber transmission began in the "O-band" just above the cutoff wavelength of the SM fiber developed to take advantage of the lower loss of the glass fiber at longer wavelengths and availability of 1310nm diode lasers.

The E-band represents the water peak region while the U/XL-band resides at the very end of the transmission window for silica glass. The E-band (water-peak band) has not yet proven useful except for CWDM. It is probably mostly used as an extension of the O-band but few applications have been proposed and it is very energy-intensive for manufacture. The E-band and U/XL-bands usually are avoided because they correspond to high transmission loss regions.

To take advantage of the lower loss at wavelength of 1550nm, fiber was then developed for the C-band. The C-band is commonly used along with the development of ultra-long distance transmission with EDFA and WDM technologies. As transmission distances became longer and fiber amplifiers began being used instead of optical-to-electronic-to-optical repeaters, the C-band became more important. With the advent of DWDM (dense wavelength-division multiplexing) which enables multiple signals to share a single fiber, the use of C-band was expanded.

With the development of fiber amplifiers (Raman and thullium-doped), DWDM system was expanded upward to the L-band, leveraging the wavelengths with the lowest attenuation rates in glass fiber as well as the possibility of optical amplification. Erbium-doped fiber amplifiers (EDFAs, which work at these wavelengths) are a key enabling technology for these systems. Because WDM systems use multiple wavelengths simultaneously, which may lead to much attenuation. Therefore optical amplification technology is introduced.

Despite great expectations, the number of installed systems using all-Raman solutions worldwide can be counted on one hand. In the future, however, the L-band will also prove to be useful. Because EDFAs are less efficient in the L-band, the use of Raman amplification technology will be re-addressed, with related pumping wavelengths close to 1485nm.

Although CWDM is now considered as a low-cost version of WDM that has been in use, most do not work over long distances. The most popular is FTTH PON system, sending signals downstream to users at 1490nm (in S-band) and using low cost 1310nm transmission upstream. Early PON systems also use 1550 downstream for TV, but that is being replaced by IPTV on the downstream digital signal at 1490nm. Other systems use a combination of S, C and L bands to carry signals because of the lower attenuation of fibers. Some systems even use lasers at 20nm spacing over the complete range of 1260nm to 1660nm but only with low water peak fibers.

Although various wavelength bands of the O-, S-, C- and L- bands have come into use with the explosive expansion of the traffic in recent years, the optical fiber amplifiers for the O- and S-band wavelengths were not realized for many years because of many technical hurdles. C- and L-band most commonly used in fiber optic networks will play more and more important roles in optical transmission system with the growth of FTTH applications.

Originally published at http://www.china-cable-suppliers.com

Installation Guide to CWDM MUX/DEMUX System

CWDM MUX/DEMUX System Overview

Coarse Wavelength Division Multiplexing (CWDM) is a wavelength multiplexing technology for access networks. It is designed to increase fiber optic network capacity without adding additional fibers. The wavelengths of CWDM channels range from 1270nm to 1610nm with 20nm spacing, which allows the use of cost-effective lasers. CWDM MUX/DEMUX system is a passive, optical solution to increase the flexibility and capacity of existing fiber lines in high-speed networks. By adding more channels into available fibers, the CWDM MUX/DEMUX system enables greater versatility for data communications in ring, point-to-point, and multipoint topologies for both enterprise and metro applications.

CWDM MUX/DEMUX System Components

All CWDM system components are passive and require no power supplies. They consist of the rack mount chassis, a set of CWDM MUX/DEMUX and CWDM OADM (Optical Add/Drop Multiplexing) modules with color-coded ports. The CWDM MUX/DEMUX takes 4 or 8 different wavelength channels and multiplexes them onto one common fiber cable for transmission to the network. Then it demultiplexes the channels it receives from the network and sends each channel to a different device. Multiple modules may be chained through the expansion port on the four-channel modules. Thus it increases flexibility and enables growth for evolving networks.

The CWDM OADM module can add or drop CWDM channels into an existing backbone ring. It provides the ability to drop one CWDM channel from the network fiber, while allowing all other channels to continue pass to other nodes. Similarly, the drop/insert module removes an individual channel from the network fiber, however, it also provides the ability to add that same channel back onto the network fiber. The drop/insert module supports two paths (east and west) for dropping and adding, so that network viability is maintained in a ring topology, even if a break occurs in the ring.

CWDM MUX/DEMUX System Installation Guide

Step1: Mount the system chassis on the rack. The CWDM rack-mount chassis can be mounted in a standard 19-inch cabinet or rack. Make sure that you install the rack-mount chassis in the same rack or an adjacent rack to your system so that you can connect all the cables between your CWDM MUX/DEMUX modules and the CWDM SFP transceivers.

Step2: Install the CWDM MUX/DEMUX modules. First loose the captive screws on the blank of module panel and remove the panel. Then align the module with the slot of the chassis shelf and gently push the module into the slot. Finally, ensure that you line up the captive screws on the module with the screw holes on the shelf and tighten them up.

Step3: Install CWDM SFP transceivers. Since each channel has a specific wavelength, transceivers must comply with the right wavelengths. Each wavelength must not appear more than once in the system. Device pairs must carry transceivers with the same wavelength.

Step4: Install the CWDM MUX/DEMUX to the switch. After inserting the CWDM SFP transceiver into the switch, single-mode patch cables are used to connect the transceiver to the CWDM MUX/DEMUX module.

Step5: Connect the CWDM MUX/DEMUX pairs. In a CWDM MUX/DEMUX system, multiplexer and demultiplexer must work in pairs. Two strands of single-mode patch cables are needed in the duplex MUX/DEMUX module and one strand for the simplex one. Simply insert single-mode cables from your system equipment to the appropriate port on the CWDM MUX/DEMUX or OADM module.

Conclusion

CWDM MUX/DEMUX system is an attractive solution for carriers who need to upgrade their networks to accommodate current or future traffic needs while minimizing the use of valuable fiber strands. With CWDM technology, you can accommodate Ethernet and SONET on a single fiber that enables converged circuit/packet networks at high demand access sites. Besides, CWDM MUX/DEMUX can work seamlessly with transceivers to optimize link length, signal integrity and network cost, thus becoming a single rack-mount solution for enhanced design, power and space efficiency.

 Source: http://www.china-cable-suppliers.com/

Higher Speed Transmission With Parallel Optics

Parallel optics is a term representing both a type of optical communication and the devices on either end of the link that transmit and receive information. Compared with traditional optical communication, parallel optical communication employs a different cabling structure for signal transmitting aiming at high-data transmission for short reach multimode fibers that are less than 300 meters. Traditional fiber optic transceivers cannot satisfy the increasing demand for high speed transmission, like 40GbE, while parallel optics technology can be a cost effective solution for 40/100GbE transmission.

Comparison between parallel optics technology and the traditional serial optical communication would better explain what parallel optics is and the reason why it is a cost effective solution to high data rate transmission. The following of this article will offer the comparison between the two optical communication technology from two aspects: connectivity method and key components.

Connectivity Method

Literally, parallel optics and serial optics transmit signals in different ways. In traditional serial optical communication, on each end of the link, there are one transmitter and one receiver. For example, the transmitter on End A communicates to the receiver on End B, sending a single stream of data over a single optical fiber. And a separate fiber is connected between the transmitter on End B and the receiver on End A. In this way, a duplex channel is achieved by two fibers.

While in parallel optical communication, duplex transmission is achieved in a different way. A signal is transmitted and received through multiple paths, thus, the parallel optical communication can support higher data rate than the traditional optical communication. This is because, the devices for parallel optic communication on either end of the link contain multiple transmitters and receivers. For instance, in 2010 IEEE 802.3ba approved the 40GBASE-SR4 physical-medium-dependent multimode parallel optical solution, which uses eight fibers to transmit four duplex channels each at 10 Gigabit Ethernet. In this case, four 10Gbps transmitters on End A communicate with four 10Gbps receivers on End B, spreading a single stream of data over four optical fibers at a total data rate of 40Gbps.

Key Components

The parallel optical communication transmitting signals over multiple fibers, which has great advantages over traditional serial optical communication. It also means that it requires different components to support its high data rate transmission.

Connector — As previously mentioned, duplex transmission in serial optical communication uses 2-fiber duplex connectors, like duplex LC connectors to link the optics with other devices, while in parallel optical communication, multi-fibers are used to reach a higher data rate. Thus, multi-fiber connectors, like 12-fiber MPO connectors are used to connect with other devices. MPO connector is one key technology support parallel optical communication. This connectivity method is showed in the following picture (Tx stands for transmit; Rx stands for receive).

Optical transceiver light source — Another complementary technology for parallel transmission is the light source of parallel optics—VCSELs (Vertical Cavity Surface Emission Lasers). Comparing with the edge-emitting semiconductor lasers in the traditional optics, VCSELs have better formed optical output which enables them to couple that energy into optical fibers more efficiently. In addition, VCSELs emit from the top surface, they may be tested while they are part of a large production batch (wafer), before they are cut into individual devices, which dramatically lowers the cost of the lasers. The following chart is about the comparison between VCSELs and edge-emitting semiconductor lasers. Cheaper to manufacture, easier to test, less electrical current required, supporting higher data rate, parallel optics using VCSELs could be a better choice to reach 40/100GbE transmission compared with traditional serial optics.

Parallel Optics for 40/100GbE Transmission

IEEE has already included physical layer specifications and management parameters for 40Gbps and 100Gbps operation over fiber optic cable. Two popular parallel optics solutions for 40Gbps and 100Gbps over multimode fibers are introduced here. For 40G, 40GBASE-SR4 transceiver is usually used, which requires a minimum of eight OM3/OM4 fibers for a transmit and receive link (4 fibers for Tx and 4 fibers for Rx). 100GBASE-SR4 transceiver (eg. QSFP-100G-SR4) is for 100Gbps transmission, which works through 4x25Gb/s 850nm VCSEL-based transmitter to achieve maximum link length of 100m on OM4 multimode fiber.

Conclusion

Parallel optical communication uses multiple paths to transmit a signal at a greater data rate than the individual electronics can support. Parallel transmission can either lower the cost of a given data rate (by using slower, less expensive optoelectronics) or enable data rates that are unattainable with traditional serial transmission. The capabilities and uses of parallel optics and MPO technology continue to evolve and take shape as higher-speed fiber optic transmission, including 40/100GbE. It is uncertain that parallel optical communication would be the trend in the future.

High-Speed Fiber Patch Cable Solutions from FS.COM

New technology advances in networking, such as 10G, 40G and 100G Ethernet solutions, mean that data center administrators face new challenges: maintaining high availability, reducing costs, seeking out higher efficiency and planning for future growth. Fiber optic patch cable, the key component in fiber optic communication system is designed to interconnect or cross connect fiber networks within structured cabling systems. It is used in data centers to interconnect ports and transceivers that accept LC and MPO/MTP fiber optic connectors. FS.COM offers a full range of cost-effective fiber patch cable solutions which can meet today and future growth. This article will demonstrate FS.COM high-speed fiber patch cable solutions for 10G, 40G, and 100G Ethernet transceiver ports interconnection as well as the cost-effective fiber patch cable solution for high-density patching.

10G Transceiver Interconnected Solution

Today’s data centers are still primarily architected around 10G Ethernet. After almost ten years of revolution, SFP+ gradually becomes the main stream of 10G transceiver in data center optics market. According to the optical ports of SFP+ form factor, the duplex LC patch cable is required to complete the link between two SFP+ modules which are plugged into switches, routers or server NICs (Network Interface Cards). FS.COM offers high quality standard LC duplex patch cables which are available in single-mode and multimode patch cable. With a wide range of material options, they can meet any working environment.

40G Transceiver Interconnected Solution

40G transceivers are ramping up hard as data centers deploy 40G Ethernet. QSFP+ transceivers, as the most popular form factor, are being widely used in data center switching fabrics. For the short reach interconnection between two QSFP+ ports, each QSFP+ transceiver requires an MPO/MTP connection. FS.COM offers MTP to MTP (or MPO to MPO) assemblies in multimode or single-mode version, with jacket ratings of riser, plenum and LSZH. With popular multimode OM3 and OM4 cable assemblies, users can easily upgrade to future 40/100G applications.

For single-mode 40G QSFP+ interconnection, it is commonly used with duplex LC single-mode patch cable. But for 40GBASE-PLRL4 QSFP+, a 12-fiber MPO/MTP single-mode cable is needed. As we know, a single QSFP+ port (4 x 10 Gbps) can breakout to four SFP+ port. Thus, using FS.COM MPO/MTP to LC assembliescan easily achieve the migration of 10G to 40G.

100G Transceiver Interconnected Solution

100G is considered the trend of the year of 2016. 100G transceivers including CXP, CFP, CFP2, CFP4 and QSFP28 are available for different applications. For applications requiring data rates of 100G, FS.COM provides multiple cost-effective solutions as follows:

• CXP/CFP to CXP/CFP Interconnection
  FS.COM’s 24-fiber MPO/MTP assemblies are ideal for 100GBASE-SR10 CXP/CFP to CXP/CFP interconnection in data center, since it is implemented 10 lanes of 10 Gbps. Among the 24 fibers, only 20 fibers in the middle of the connector are used to transmit and receive at 10 Gbps and the 2 top and bottom fibers on the left and right are unused.
• QSFP28 to QSFP28 Interconnection
  The QSFP28 is the exact same footprint as the 40G QSFP+, but is implemented with four 25Gbps lanes. To interconnect a multimode QSFP28 link, a 12-fiber MPO/MTP patch cable is required, while for single-mode link (100GBASE-LR4 QSFP28), a duplex LC single-mode patch cable is required. The interconnection of QSFP28 multimode link is similar with the case of 40GBASE-SR4 QSFP+.
• CXP/CFP to 10 x SFP+ Interconnection
  As mentioned above, 100GBASE-SR10 CXP/CFP uses ten 10Gbps lanes to achieve 100Gbps data rate. Thus, a CXP/CFP port can be breakout to ten SFP+ ports using 24-fiber to LC harness cables from FS.COM.

High-Density Patching Solution

High-density patch cables are with higher performance and easier to use compared to common ones. FS.COM offers a wide range of high-density fiber patch cables, such as LC-HD and MPO-HD TAB fiber patch cable, Keyed LC fiber patch cables, LC uniboot fiber patch cables, bend insensitive fiber patch cables, etc. All these patch cables are with high performance and can be customized according to specific requirements.

Direct Uplink vs. Fan-Out Uplink

Fan-out cable or breakout cable is considered as one of the the latest enabling technologies to help increase port densities and lower costs. Taking one (large bandwidth) physical interface and breaking it out into several (smaller bandwidth) interfaces, it has been highly recommended to be used in network migration. In this post, I will compare the 40G direct uplink and the fan-out 4x10G uplink configurations and their inherent differences in maximum server scalability.

For a leaf/spine network architecture, the number of connections used for uplinks from each leaf switch determines the total number of spine switches in the design. Meanwhile, the number of ports on each spine switch determines the total number of leaf switches.

Now, imagine that we’re building a network that supports 1200 10G servers in one fabric with 2.5:1 oversubscription. The network must seamlessly expand to over 5000 10G servers without increasing latency or oversubscription in the future. Next, I will compare two different approaches.

40G Leaf/Spine Fabric – With 40G QSFP Uplink

The following picture shows us a 40G fabric to us. Each of the spine switches has 32 ports of 40G. Each leaf ToR (Top of Rack) switch is connected to the spine with 4 ports of 40G using the leaf switches 40G QSFP uplink ports. There will be 40 servers per rack connected to each leaf switch. Namely, it allows a maximum is 1280 x 10G servers at 2.5:1 oversubscription. This meets the initial scale target of 1200 servers however, it cannot scale larger. Before the network can achieve the 5000 server design goal, the 40G design will have to be re-architected.

10G Leaf/Spine Fabric

As mentioned above, a direct 40G uplink is not an ideal configuration for a cost-effective server scalability. So, what level of server scalability can be achieved when using fan-out the leaf switch uplink ports with 4 x 10 G each?

The picture below shows a 10G leaf/spine fabric to us. With a QSFP+ to SFP+ optical fan-out cable, each QSFP leaf port is now fanned out to 4 x 10G interfaces each, for a total of 16 x 10G uplinks. Each of the spine switches now has 128 ports of 10G. Each leaf switch is connected to the spine with 16 ports of 10G. In this case, the maximum scale is 5120 x 10G servers at 2.5:1 oversubscription. Obviously, with the same bandwidth, latency and oversubscription, this fabric is better. It can not only be built to 1200 servers for present demand but also can seamlessly scale to over 5000 servers in the future.

In conclusion, the 10G leaf/spine fabric design offers 4X greater scalability compared to the 40G fabric design with the same hardware, bandwidth, latency and oversubscription. These two configuration scenarios show us how fan-out technology are used to scale up a data center fabric. As the 10G network is widely deployed in today’s data center, the 10G direct attach cables (like Cisco SFP-H10GB-ACU10M) are preferred by many designers and companies. But higher speed demand such 40/100G is also needed. The fan-out technology not only enables new levels of server scalability, but also promotes a time-saving and cost-effective network migration.