Brief Introduction on 25G/50G/100G Ethernet

The rise of cloud computing and the expansion of the data center are pushing the latest Ethernet speeds up, while big data based on cloud technology has already added to carriers’ workloads. To meet this requirement, the data center extends the bandwidth capabilities that are parallel to existing infrastructure. Rapid growth in the expected 25G and 100G Ethernet deployments is a testament to this trend.

In order to be able to handle the increasing data load, the industry’s largest long-distance cloud companies have together with their core network’s data center operators, to jointly use the 100G Ethernet architecture. However, most operators believe that 100G or even 40G is somewhat excessive for server connections because its workload only needs to be incrementally improved over 10G networks. This is why, although 40G and 100G Ethernet have been introduced, 25G and 50G Ethernet are still one of the reasons for the common choices within the data center. Below we will briefly explain why 25G is more suitable for these applications than 40G.

Several recent Ethernet bandwidth technologies are not designed to set a new high speed, but more to push such network protocols into adjacent markets, especially the data center market. Below we will explain the specific reasons by introducing 25G, 50G and 100G respectively.

25G

The official draft of the IEEE 802.3 draft standard for 25G Ethernet will eventually be completed in 2016, and it will mainly be aimed at servers in cloud data centers. This is a relatively short time frame due to the reusable components of 10G and 100G Ethernet.

40G and 100G already exist, but why use 25G? This confused some operators. The answer lies in the requirements of architecture and performance. The existing 100G standard network system consists of four links, each of which has a bandwidth of 25 Gbps. This four-to-one ratio is equivalent to connecting servers to 25G switches and then converging to 100G uplinks, which helps network operators to more easily expand their data centers.

Similarly, 40G Ethernet is composed of four 10G Ethernet links. However, according to John D’Ambrosia, chairman of the Ethernet Alliance, many data centers have adopted more than 10G servers. This is why a number of chip vendors have provided 25G serial/deserializer transceivers. This will not only make bandwidth convergence for 25G, 50G, and 100G Ethernet more convenient, but also reduce costs due to volume.

50G

Although the implementation of the IEEE standard for 50G Ethernet is still some time away (approximately 2018 to 2020), many industry alliances expect that products will begin to appear in 2016. Similar to 25G technology, 50G Ethernet technology will be the next solution for high-speed connection servers and data centers. According to analyst firm’s Dell’Oro data, over the next few years, servers and high-performance flash storage systems will need to exceed 25G.

To help deliver these accelerated Ethernet technology products faster, the 25G/50G Ethernet Alliance has eliminated the royalty fees for the 25G and 50G Ethernet specifications and is open to all data center ecosystem vendors.

 

Reusing the 25G components of the existing 100G network can reduce the implementation cost of 50G. For example, the cost structure of 25G cabling is the same as 10G, but its performance is 2.5 times. Similarly, 50G costs half of the 40G cost, but performance can be increased by 25%.

100G

For long-distance carrier networks ranging from hundreds of kilometers to tens of thousands of kilometers, the deployment of 100G Ethernet will continue to grow.

But according to information provided by a new industry alliance, the 100G architecture will be another excellent market alternative. The 100GCLR4 alliance led by Intel and Arista Network believes that 100G is ideal for connecting large “ultra-large” data centers spanning 100 meters to 2 kilometers.

Other companies are also seeking alternative 100G implementations for the data center. Sinovo Telecoms has joined the CWDM4MSA industry consortium, which aims to define a common specification for low-cost 100G optical interfaces within 2 kilometers of data center applications. With the transformation of network infrastructure to 100G data rates, data centers will require long-range, high-density, 100G embedded optical connections. The MSA uses Coarse Wavelength Division Multiplexing (CWDM) technology to provide four 25G single-mode fiber (SMF) link channels. Similarly, the OpenOpTIcsMSA organization initiated by Ranovus and Mellanox Technologies will also focus on developing a data center supporting 2 kilometers of 100G.

In the past, the increase in speed has driven the development of most network components. Today, to handle the massive data flow through the cloud, companies need to seek a balance between speed-up and reuse technology to find a cost-effective solution. Gigalight, as a professional optical transceiver vendor, can provide various kinds of optical transceivers to meet your 25G/50G/100G/200G/400G transmission needs. For more details, please visit its official website.






Introduction on 5 Kinds of 40G QSFP+ Optical Transceivers

40G optical transceivers are a series of optical transceivers with 40Gbps transmission rate, in which CFP and QSFP are the main form factors. And the 40G QSFP+ optical transceivers are one of the most widely used optical transceivers. In the post, Gigalight will introduce you several kinds of most popular 40G QSFP+ optical transceivers that can help you have a better choice.

40G LR4 QSFP+

The 40G LR4 QSFP+ optical transceiver is typically used with LC single-mode fiber patch cables for transmission distances up to 10km, and it has 4 data channels that transmit data simultaneously. The advantages of 40G LR4 QSFP+ optical transceivers are high density, low cost, high speed, large capacity, and low power consumption.

The working principle of 40G LR4 QSFP+ optical transceiver: The laser driver controls the arrival wavelength, the optical signal passes through the multiplexer and is combined together for transmission. When arriving at the receiving end, these transmitted signals are then demultiplexed by the demultiplexer into four channels with a transmission rate of 10 Gbps. Then the PIN detector or transimpedance amplifier recovers the data stream and transmits the optical signal.

40G SR4 QSFP+

The 40G SR4 QSFP+ optical transceivers are often used with the MPO/MTP connector in 40G data transmission. It has four independent full-duplex channels and is also transmitted through four channels. The transmission rate is the same as that of the LR4. The difference is that the 40 G SR4 QSFP+ optical transceivers are often used with multimode optical fiber. The transmission distance when using it with the OM3 optical fiber jumper is 100m, and the transmission distance when using it with the OM4 optical fiber jumper is 150m.

The working principle of the 40GBASE-SR4 optical transceiver: when the transmitting end transmits a signal, the electrical signal is first converted into an optical signal via a laser array. When the transmitting end transmits a signal, the photodetector array converts the parallel optical signal when the receiving end receives a signal.

40G LR4 PSM

As a highly integrated 4-channel optical transceiver, the 40 G LR4 PSM optical transceivers have the advantages of high port density and low cost. The optical port of this optical transceiver adopts the parallel single-mode technology-PSM. It uses a 4-way parallel design MPO/MTP interface and the transmission distance is 10km.

The work principle of 40 G LR4 PSM optical transceivers is in the same way as the 40 G SR4 QSFP+ optical transceivers. The difference is that 40G LR4 PSM optical transceivers are often used to connect with single-mode ribbon fiber connectors. That is, parallel optical signals are sent in parallel through eight single-mode optical fibers.

40G QSFP+ DAC

The 40G QSFP+DAC high-speed cable consists of two 40G QSFP+ optical transceivers and a copper cable.

DAC Advantages:

(1) Low cost, reducing the impact of dust and other contaminants on optical cables, and improving transmission efficiency;

(2) The high-speed cable is made of copper core, which has good heat dissipation effect and is energy-saving and environmentally friendly.

(3) High-speed cables consume low power.

40G QSFP+ AOC

The 40G QSFP+AOC active optical cable is the core component of the parallel optical interconnection. It is composed of two 40G QSFP+ optical transceivers connected by a ribbon optical cable.

The QSFP+AOC active optical cable is an efficient integrated fiber optic cable assembly designed for short-range, multi-channel data communication and interconnection applications. Each signal direction has four data channels with a rate of 10 Gbps per channel.

 AOC Advantages:

(1) The transmission power is lower, so the power consumption is small;

(2) Weight and volume are much smaller than high-speed cables;

(3) The transmission distance is farther (it can reach 100-300 meters).

In Conclusion

The above five kinds of optical transceivers are available from Gigalight. When you use the optical transceiver purchased in Gigalight, your device’s stability and network speed will be greatly improved. If you want to learn more about optical transceiver solutions, please visit Gigalight official website to view.






10G SFP+ DAC vs. 10G SFP+ Transceivers

The development of artificial intelligence and Internet of things presents new challenges to the expansion of data centers, and there is often a contradiction between technology and cost. In order to realize high density and high capacity, it is important to control cost factors and reasonable wiring. In the wiring, we can choose the high-speed cable and the optical transceiver cables, so how do we choose these two products in the actual scene? What are the differences and what advantages do they have? Let’s study together about the differences between 10G SFP+ DAC and 10G SFP+ transceivers.

As a transmission medium, 10G SFP+ DAC and 10G optical transceivers can be selected. What is the difference between the two?

  • The 10G DAC is connected to two switches through copper cables. The SFP+ optical transceiver is connected to the jumpers to connect the two switches.
  • 10G DAC is short-distance transmission; the longest distance is 15M, used in the engine room.
  • The SFP+ transceiver can perform long-distance transmission. The longest single fiber is 80KM, and the longest dual fiber is 100KM.

The Advantages of 10G SFP+ DAC:

The 10G DAC is a copper cable designed with SFP+ connectors on both ends and is less expensive than a 10G optical transceiver.

The use of 10G DAC wiring is more flexible, transmission distance up to 15 meters, in the actual construction process is less difficult to operate.

10G DAC cabling saves on connected devices, eliminating the need for patch panels, and servers and network equipment can be directly connected to TOR switches, which indirectly save on input costs.

The Advantage of 10G SFP+ Transceivers:

If the vertical distance of the wiring does not exceed the cabinet, 10G DACs can be used for the connection. When the distance between the TOR switch and the network switch is greater than 15M, multimode optical fibers and fiber transceivers can be selected. Usually, OM3/OM4 LC fiber jumpers and 10G SFP+ optical transceivers are used. In other words, 10G SFP+ optical transceivers are widely used in long-distance transmission.
Gigalight provides high-speed direct connection solutions for data center interconnection, including 10G SFP+ to SFP+ high-speed cable solutions, which not only reduces power consumption, but also increases network scalability. Want to learn more about the product details? You can visit our website.






Tips on How to Use Optical Transceivers

Optical transceiver consists of optoelectronic devices, functional circuits, and optical interfaces. The optoelectronic devices include transmit and receive parts. The transmitting part is: Inputting a certain bit rate of the electric signal is processed by an internal driver chip to drive a semiconductor laser (LD) or a light emitting diode (LED) to emit a corresponding rate of modulated optical signal, and an internal optical power automatic control circuit is provided therein. The output optical signal power remains stable. The receiving part is: After a certain code rate of the optical signal input transceiver is converted into an electrical signal by the light detecting diode. After the preamplifier outputs the corresponding rate of the electrical signal, the output signal is generally PECL level. At the same time, an alarm signal will be output after the input optical power is less than a certain value.

Today Gigalight will share with everyone some tips on using optical transceivers if you usually pay attention to the maintenance of the optical transceiver. Note that the following two points can help you reduce the loss of the optical transceiver and improve the performance of the optical transceiver.

Note One:

1. There are CMOS devices in this chip. Pay attention to prevent static electricity during transportation and use.

2. The device grounding should be good, reduce parasitic inductance.

3. As far as possible manual welding, if you need to paste, control the reflow temperature cannot exceed 205℃.

4. Do not lay copper below the optical transceiver to prevent the impedance from changing.

5. The antenna should be away from other circuits to prevent radiation efficiency becomes lower or affect the normal use of other circuits.

6. The transceiver should be placed as far away from other low-frequency circuits, digital circuits.

7. It is recommended to use magnetic beads for the isolation power of the transceiver.

Note Two:

1. Do not look directly into the optical transceiver that has been inserted into the device (whether it is a long-range or short-range optical transceiver) with naked eyes, and avoid eye burns.

2. When using a long-distance optical transceiver, the transmit optical power is generally greater than the overload optical power. Therefore, it is necessary to pay attention to the length of the optical fiber and ensure that the actual received optical power is less than the overload optical power. If the length of the optical fiber is short, use a long-range optical transceiver and use it with light attenuation. Be careful not to burn out the optical transceiver.

3. To better protect the optical transceiver from cleaning, it is recommended that you plug the dust plug when it is not in use. If the optical contact is not clean, it may affect the signal quality, it may also lead to link problems and error codes.

4. Rx/Tx, or arrow in and out directions is generally marked on the optical transceiver to facilitate identification of the transceiver. Tx at one end must be connected to Rx at the other end, otherwise the two ends cannot be linked.

Read the above notes, whether do you have a new understanding of the use of optical transceivers? It is important to be helpful to everyone and thank you for your support and attention to Gigalight. For more product details, please visit our official website.

What Can Pluggable Optical Transceivers Do in Data Centers

For data centers, fiber-optic technology is no longer an option, or is only used to solve the most difficult interconnection problems. Today, high broadband, high port density and fiber optic technology are needed to solve low power requirements, and the current optical fiber technology is a kind of batch production technology, low cost, and is widely used in various applications such as switches interconnect and server interface. And in this post, Gigalight will introduce what pluggable optical transceivers can do in data center in detail.

1. Extend Data Center Distance

From 100Mb/s to 100Gb/s, single-channel 25G Ethernet optical transceivers lead the optical transceiver market of next-generation servers and switches. 40G QSFP+ products can support transmission distances up to 300m over multimode optical fibers, which greatly exceeds the standard distance of IEEE 40G Ethernet. In the 40G QSFP+ that transmits on single-mode fiber, and the 10 GSFP+ product that transmits 80 km, our OIF module or CFP2-ACO module supports a transmission distance of 500km or more for data center metro or intercity connectivity.

2. Increase Density and Reduce Power Consumption

Our products are at the leading edge of the next generation of low-power optical transceiver products. The 100G QSFP28 optical transceivers (SR4, LR, CWDM4, and SDWM4) have a maximum power consumption of only 3.5W. The 40G and 100G Quad wire active fiber optic products have power settings that can be flexibly configured by the host system.

3. Deploy with Existing Multimode Fiber

Most data centers today are still based on the 10G Ethernet architecture and use 10Gbase-SR short-haul transmission over OM3/OM4 duplex multimode fiber. With the data center upgrading from 10G to 40G or even 100G, customers still want to retain the existing multi-mode fiber architecture. However, SR4 optical transceivers need to be connected with ribbon multimode optical cables (multi-core) on the interface, and LR4 optical transceivers need to be double. Single-mode fiber, both of which are not present in the data center of currently deployed duplex multimode fiber, QSFP+ LM4 modules allow customers to implement 40 links over existing duplex multimode fiber, SWDM4 modules for customers in the the existing affordable dual-mode multimode fiber architecture enables 40G and 100G Ethernet transmission solutions.

4. from 100G to 200G/400G

Since 2010, the 100G Ethernet optical transceiver has been in a leading position in the market, supplying a large number of CFP optical transceivers for the operator’s routers and transmission systems. Since then, we have continued to expand 100G products and developed and supplied CFP2, CFP4, Modules such as CXP and 100G QSFP28 should be widely used in telecommunication, emerging data centers, and 100G systems in enterprise networks. However, we have not stopped our steps. At present, we are actively leading the development of industry standards and the development of next-generation Ethernet products, including 200G and 400G-rate products that will meet the long-term technical requirements for future high-performance data centers.

Three Trends to Drive 100G Ethernet Development

The Ethernet market has seen tremendous growth over the past few years. Accelerating the transmission speed and expanding the capacity of the data center will help promote this trend. According to the IHS Infonetics report, by 2019, 100 Gigabits per second 100G Ethernet will account for more than 50% of the data center fiber transceiver market. As 100G chips are being put into production, the market for 100G Ethernet is accelerating. In this article Gigalight will analyze the three major trends driving the development of the 100G Ethernet market.

1. Data Center Architecture and Traffic Changes

At present, the transmission technology of the optical fiber industry reaches gigabits per second (10G) and 40 gigabits per second (40G), which has been a long time. These technologies are effective and most people have no objection to this. For most users, the 40G transmission speed is more than enough. The problem of data transmission in the data center becomes obvious. The scale and traffic of Internet content providers and companies on cloud data will continue to grow.

Cisco Systems predicts that global data center Internet Protocol (IP) traffic will grow at a 31% annual rate over the next five years. Changes in the way people use the Internet have contributed to this growth. The amount of data in cloud computing is getting larger and larger, and more and more data are being accessed by mobile devices in the world to access video social media content.

The construction of data centers is increasing, which requires a better data management solution. Influx of traffic has led to changes in the way three-tier networks and other ways of changing the flow of information through the data center (that is, the user interface, data processor, and database management system combined). Newer technologies allow parallel processing and can transfer more data. The Internet is becoming more and more complex and websites need more interconnections. The architecture of the data center is changing, focusing more on integrating nodes and increasing bandwidth speed. It is clear that 100G will become the new standard for higher bandwidth and smarter data center architectures.

2. 10G Can’t Meet the Growing Demand for Corporate Networks

Some large data centers have switched. The Howard Hughes Medical Institute recently switched to 100G technologies, delivered through the Brocade MLXE router. The data center includes 56 11G ports, all equipped, and its efficiency has reached the highest priority of the switch. Traditionally, data centers will rely on 10G multi-beam transmissions that require link aggregation and lead to sub-optimal and inefficient load balancing.

This is where 100G comes into play. It frees up space, minimizes data aggregation, and significantly increases overall efficiency. As companies continue to grow in size and data needs become more complex, 100G will provide them with the bandwidth speed and efficiency they desperately need. Companies with four or five 10G ports have witnessed their database growth and may find switching to more affordable and scalable 100G ports. Of course, this is driven by costs and the resources of the company.

3. The Continuous Development of CMOS Technology Will Make 100G Become Mainstreams

With the evolution of 10G technology, before 100G technology becomes mainstream, which requires a certain amount of time to develop transceiver technology. When it is adopted, it is expensive and requires a lot of power. Over time, advances in chip technology have reduced more costs and various energy-saving technologies have emerged. This is exactly the reason that 100G technologies wins in the market, and the adoption of CMOS technology makes it an industry standard. Because using CMOS architecture will make it faster and use less power at the same time.

Once the technology is mature, the 100G system architecture can save more power and provide up to 10 times the speed. Currently, Cisco and Brocade Communications Systems sell 100G switches and routers at the enterprise level. However, the average cost per port of the switch is 2,500 US dollars, which means that companies using 100G networks will have to pay a lot of money. However, with the development of CMOS technology, creating these systems will become easier and more economical. These systems will reduce costs, reduce data center size and power requirements, and make 100G applications mainstream.

About Gigalight:

Gigalight is a design innovator in global optical interconnect field. A series of optical interconnect products include: optical transceivers, passive optical components, active optical cables, GIGAC MTP/MPO cabling, cloud programmers & checkers, etc. Three applications are mainly covered: Data Center & Cloud Computing, MAN & Broadcast Video, and Mobile Network & 5G Optical Transmission. Gigalight takes advantage of its exclusive design to provide clients with one-stop optical network devices and cost-effective products.

What Is CSFP (Compact SFP)

Compact SFP is a new kind of fiber transceiver usually known as CSFP fiber optic transceivers. CSFP is short for Compact SFP. CSFP has the same size of SFP, Cisco also called this as 2- channel bi-directional SFP. And in this post, we will guide you have a deeper understanding about Compact SFP if you are interested.

Introduce CSFP MSA and CSFF MSA:

The CSFP MSA defines a transceiver mechanical form-factor with latching mechanism and a host board, SFP-like, electrical edge connector and cage. The CSFF MSA also defines a transceiver mechanical form-factor. The Dual-Channel CSFP has the same mechanical dimensions as the industry standard SFP transceiver and is compatible with the standard SFP cage. The Single-Channel CSFP and CSFF are half the size of the industry standard SFP and SFF packages. The CSFF design is modular to enable configurations of integrated 1, 2 or 4 channel modules. These highly integrated compact transceiver modules will enable network system vendors to increase port density and data throughput, while reducing network equipment cost. SOPTO CSFP transceivers are compatible with the Compact Small Form- Factor Pluggable (CSFP) Multi-Source Agreement (MSA).

What Is CSFP?

CSFP (Compact Small Form Factor Pluggable) is a Compact SFP, which develops a more advanced and compact CSFP package based on the current popular SFP package. By adopting dual channel, the design of the four channels, CSFP uses the existing SFP common interface, but reducing the overall dimensions of existing half of the industry standard and a quarter. By combining channel number, it can also be flexible configuration. If the traditional discrete element scheme is continued, it will be difficult to achieve the above functions in technology. CSFP combines with high integration of optoelectronic integrated technology, on the basis of all the technical advantages of a SFP, greatly decreasing the size of the shape of the optical transceiver modules and optical system, a significant increase in density of communication port and data throughput, reduce the system cost. It is expected to play a prominent role in the data communications market. CSFP has 155M, 1.25G and 2.5G in speed.

When Compact SFP (CSFP) Is Used?

Most commonly CSFP is used in Central Office/Aggregation site operating at Tx1&Tx2:1490nm Rx1&Rx2:1310nm and it get’s connected to 2 CPE sites which have simple BiDi SFP operating in Tx:1310nm Rx:1490 at each site. Compact SFP (CSFP) major application is to use it in FTTx scenario where we have (point-to-point) connections from Central Office/Aggregation Site to (CPE) Customer Premises Equipment. CSFP can help us double the Central Office/Aggregation Site port density (we can achieve 2xGE from one aggregation site port) and as well reduce power consumption in Central Office as we use 2 times fewer ports.

How about Gigalight CSFP?

The Gigalight CSFP is a series of optical transceiver modules operating at over dual Single Mode Fiber (SMF) as a 2-channel BiDi SFP. These optical transceivers are designed for use in Fast Ethernet, Gigabit Ethernet, Fibre Channel, and SONET/SDH links, compliant with the SFP MSA. Digital diagnostics functions are available. The optical transceivers are RoHS compliant. For more details, you can visit our official website to know more.

 

Single-wave 100G Optical Transceivers for Upgrade of Data Centers

Data center industry is facing unprecedented development opportunity; the development of various applications of network gave rise to the vast amounts of data requirements. The artificial intelligence has brought high density calculation. All of the information technology has led to the data center to develop to a higher density, higher bandwidth.

The number of QSFP28 optical transceivers used for optical interconnection in data centers only in 2017 has reached nearly 3 million, and the annual compound growth rate of over 100 percent has continued to increase to 2021.

100G QSFP28 optical transceiver series products for data centers have entered the technology maturity, and rapidly develop in the second half of 2016, among which the fastest growing is the QSFP28 optical transceivers used in the data center interconnection.

Single-wave 100G QSFP28 DR1/FR1 Optical Transceivers Have Become Popular Solutions

At the current market, the optical transceivers based on 4*25G NRZ technology are mainly used for the 100G solution. This technology has a wide range of products in transmission distances: 100m SR4/SWDM4, 300m eSR4, 500m PSM4/CWDM4 Lite, and 2km CLR4/CWDM4. These different types of products create great difficulties for the interconnection, interoperability and compatibility of the Internet, thereby increasing the cost of data centers. Therefore, single-wave 100G optical transceivers are increasingly concerned by Internet companies. According to IEEE regulations, DR1/FR1 is based on the PAM4 modulation format and supports a single-channel 100Gbps (53GBd) class of physical media dependence. DR1 can support single-mode fiber transmission at least 500m; FR1 supports 2km transmission. Compared with the current 100G QSFP28 optical transceivers (4*25Gbps), DR1/FR1 uses only one optical device, which can greatly save the cost of optical transceivers. The DR1/FR1 optical transceivers based on the QSFP28 package can be completely compatible with the existing 100G switch.

Single-wave 100G Optical Transceivers: the Best Solution for the Next Generation of Data Center

From the product application level, 100G QSFP28 DR1/FR1 optical transceiver can be considered as the solution for the next-generation 100G QSFP28 optical transceivers, and they can cover the transmission distance of the vast majority of data center optical interconnections. And it is expected to replace all current QSFP28 optical transceiver solutions within 500m, which effectively avoids the problem of 100G optical transceiver solution selection and interconnection. Another important application of DR1 is that it can be used for fan-out connection with DR4, between the top switch and the server. From the perspective of the evolution of technology R&D, DR1/FR1 can be used as the basis for research on LR1 (electrical solutions are almost the same), and it can also accumulate corresponding data for the development of DR4/FR4 (optoelectronic solutions are all the same, only the expansion of channels).

From the technical level, 100G QSFP28 DR1/FR1 technology uses the PAM4 modulation technology, which greatly improves the signal rate and can effectively reduce the number of signal channels compared to the NRZ modulation technology. The PAM4 modulation technology replaces the existing NRZ modulation technology in the next-generation 400G optical transceiver products, reducing the size of the optical transceivers and reducing power consumption, thereby reducing the cost of the optical transceivers.

Technologies of Data Center Optical Transceivers

With the commercial use of cloud computing, big data and other new technologies, data center flow and bandwidth have an index incensement. According to the LightCounting forecast, by 2019, the sales volume of data center optical transceivers will be over $50 million and the market scale is hopeful to reach $4.9 billion, which will be a huge opportunity for optical transceiver vendors. At the same time, we can see that there are some difference in applications of optical transceivers between data center and Telecom. In the post, we will discuss the technology development direction of data center optical transceivers in detail.

On the macro level, the data center optical transceiver market is a market that reasonably defines the life and working conditions of optical transceivers according to the actual requirements, and fully optimizes the market for the cost performance of optical modules. Due to the open trend of several networks, this market has the characteristics of positivity and open, welcoming the characteristics of new technologies and the atmosphere of exploring new standards as well as application conditions. All of these provide excellent conditions for the development of data center optical transceiver technology. Here we are trying to enumerate the development direction of some data center optical transceiver technology for your reference.

Non-hermetic Package

As the cost of optical components (OSA) accounts for over 60% of the cost of optical transceivers, and the space for cost reduction of optical chips becomes smaller and smaller, the most likely cost reduction is the packaging cost. While ensuring the performance and reliability of optical transceivers, it is necessary to promote the packaging technology from the expensive hermetic package to the low-cost non-hermetic package. The key points of the non-hermetic package include the non-air tightness of the optical device itself, the optimization of the design of the optical components, the packaging materials and the improvement of the process. Among them, optical devices, especially lasers, are the most challenging. This is because if the laser device is not hermetic, expensive hermetic package is not needed. Fortunately, in recent years several laser manufacturers have avowed that their lasers can be applied to non-airtight applications. In view of the large number of shipping data center optical transceivers, most of them are mainly non-hermetic package. It seems that the non-hermetic packaging technology has been well received by the data center optical transceiver industry and customers.

Hybrid and Integrated Technology

Under the drive of multi-channel, high speed and low power consumption demand, the same volume optical transceivers need to have more data transmission, and the photonic integration technology gradually becomes a reality. Photonic integration technology has a broader meaning: for example, based on the integration of silicon-based (planar optical waveguide hybrid integration, silicon photonics, etc.), based on the integration of indium phosphide. The hybrid and integrated technology usually refers to the integration of different materials. There are also the construction of partially free space optics and partially integrated optics called hybrid integration. The typical hybrid integrated active optical devices (laser, detector, etc.) are integrated into the passive optical path connection or some other function (points or wave, etc.) of the substrate (planar optical waveguide and silicon light, etc.). Hybrid integrated technology of optical components can be done very compactly, complying with the trend of miniaturization of optical transceivers, easy to use mature IC encapsulation process automation. It is beneficial to mass production, which is an effective technical method for recent data center optical transceivers.

Flip Chip Technology

Flip chip is a high-density chip interconnection technology from IC packaging industry. In the rapid development of optical transceivers today, the interconnection between short – shrinking chips is a valid option. It is better to weld optical chip directly onto the substrate through gold-gold welding or eutectic welding, which is much better than the high frequency effect of gold wire bonding (short distance, small resistance, etc.). In addition to the laser, the heat generated by the laser is easily transferred from the solder to the substrate due to the proximity of the source area to the solder, which is helpful for improving the efficiency of the laser at high temperature. Because the backward welding is the mature technology of IC packaging industry, there are many kinds of commercial automatic reverse welding machines used in IC packaging. Optical components require optical path coupling, so the accuracy requirements are high. These years optical components processing with high precision inversion welding machine are very eye-catching and in many cases have realized the passive light, greatly improving the productivity. Due to the characteristics of high precision, high efficiency and high quality, the flip chip technology has become an important technology in the data center optical module industry.

Chip On Board Technology

COB (chip on board) technology also comes from the IC packaging industry, whose principle is through the rubber patch technology (epoxy die bonding) to fix chips or optical components on the PCB, and then gold wire bonding (wire bonding) uses electrical connection, and lastly drip glue sealing on the top. Obviously, this is a non-hermetic package. The advantage of this process is that it can be automated. For example, the optical components can be viewed as a “chip” after it has been integrated by back loading and welding. Then the COB technology is used to fix it on the PCB. At present, COB technology has been widely adopted, especially in the use of VCSEL arrays in short distance data communication. The integrated silicon photonics can also be packaged by using COB technology.

Silicon Photonics Technology

Thh silicon photonics is a technology that discusses the technology and technique of optoelectronic devices and silicon-based integrated circuits, and a science integrated into on the same silicon substrate. Silicon photonics technology will eventually go to photoelectric integration (OEIC: Opto – Electric Integrated Circuits), making the current separated photoelectric conversion (optical transceivers) into local photoelectric conversion of photoelectric integration, further pushing the system integration. Silicon photonic technology can certainly do a lot of things, but for now it’s the silicon modulator. From the industry, the threshold of a new technology into the market must be the performance and cost is competitive and the need for huge upfront costs of silicon photonics technology is really a big challenge. The data center optical transceiver market, due to the large demand concentration within 2 kilometers, with the strong requirements of low cost, high speed and high density, is suitable for a large number of applications of silicon photonics.

In my opinion, the traditional 100G optical transceivers have been very successful, and they are not easy to get a lot of silicon photons. However, at the rate of 200G or 400G, since the traditional direct modulation type is close to the limit of bandwidth, the cost of EML is relatively high, which will be a good opportunity for the silicon photonics. The large number of applications of silicon photons also depends on the openness and acceptance of technology in the industry. If taking into account the characteristics of silicon photonics when setting the standards and agreements or relaxing some indicators (wavelength, extinction ratio, etc.) on the premise of meeting the transmission condition, they will greatly promote the development and application of silicon photonics.

On Board Optics

If OEIC is the ultimate photoelectric integration scheme, on board optics is a technology between OEIC and optical transceivers. On board optics moves the photoelectric conversion function from the panel to the motherboard processor or to the associated electrical chip. By saving space and increasing the density, it also reduces the distance of the high frequency signal, thus reducing the power consumption. On board optics is primarily focused on the short-range multimode fiber used in the VCSEL array, but recently there is a scheme for using silicon photonic technology in single-mode fiber. In addition to the composition of the simple photoelectric conversion function, there are also the forms (co-package) that encapsulate the photoelectric conversion function (I/O) and the associated electrical chip (processing). Although on board optics has the advantages of high density, the manufacturing, installation and maintenance costs are relatively high, and are currently used in the field of supercomputing. It is believed that with the development of technology and the need of the market, on board optics will gradually enter into the field of optical interconnection of data center.

Optics Products Are Needed in 5G Fronthaul

In the past few years, Telecom operators have already upgraded their LTE networks by using additional spectrum, carrier aggregation and LTE-A, and have added Small Cell in Macrocell coverage area to drive the increasement of fronthaul bandwidth requirements. In the current, many operators and equipment vendors have standardized the multi-rate transceivers that support 10Gb/s for all fronthaul requirements. Because they are able to meet most of different transmission speed requirement by one device and decrease the complexity of the specific site design and spare part inventory. Many operators, especially those that lease their fronthaul fiber, also deployed WDM system in their fronthaul networks.

5G Fronthaul Will Need Faster Optical Products, But How Fast?

With the emergence of 5G mobile network, the fronthaul demand will also change. The target peak bandwidth of 5G is 20Gb/s, which will require a higher spectrum than LTE to realize the requirement. That is to say, the shorter wavelength can realize the smaller antenna in the millimeter wave range, thus allows the use of higher order MIMO antenna arrays. In LTE area, 4*8 and 8*8 MIMO have been top. But in 5G area, 64*64 MIMO is also possible. The number of MIMO is higher; the bandwidth required for the corresponding fronthaul link is larger. In the case that other conditions are same, the second way for 5G to increase bandwidth is to use 100 GHz frequency (LTE uses 20GHz), so that can produce a single radio transmission from cellular site to the core network for more than 5 times of bandwidth.

Given the fronthaul bandwidth required to support 5G radio may be have a substantial growth, mobile device manufacturers update the CPRI specification to “eCPRI” (released in August 2017). One key factor of eCPRI is to transfer some physical layer signal processing from the baseband unit to the radiofrequency pull head (RRH), which in many cases reduces the fronthaul bandwidth to one in ten.

When all the different factors that influence the bandwidth of the 5G fronthaul add up (some drive its growth, some drive it down), the expected bandwidth fall in the 14 Gb/s to 30 Gb/s range, depending on the eCPRI implementation details, base stations, and etc. If the old CPRI scheme is adopted, all physical layer signal processing will remain in the baseband unit, and similar 5G network configuration will require 236Gb/s fronthaul bandwidth. As a result, the 5G base station will generate 160Gb/s or more in nominal terms, and with eCPRI, the actual fronthaul bandwidth required will be 14-30Gb/s.

Just like that 10G optical transceivers can become the actual standard for LTE fronthaul, the next generation of higher standard Ethernet speeds will be applied to 5G fronthaul, which means that 5G deployment will require a large number of 25GbE devices. Even though some components are industrial temperature and/or bidirectional versions specially designed for fronthaul application.

5G Network Will Also Need Higher-speed Optics Products( 25Gb/s or above)

Mobile fronthaul or backhaul need 50G, 100G or even 400G optical transceivers. CPRI alliance has defined fronthaul for a long time, but there is no a consistent definition for wireless backhaul. LightCounting defines the backhaul as the first optical link that begins in BBU and carries the flow from the core network. Other broader definitions include access, aggregation, and core networks. Naturally, if the data flow from BBU to the core network flows to 25Gb/s, then 50G, 10G, or even 400Gb/s transfers may be needed.