Higher bandwidth requirements and faster speed are enhancing the need for 400G Ethernet and optical modules in the large data center interconnections. Multiple network equipment manufacturers, cloud service providers, and optical transceiver vendors have raced to the 400G market in a quick and furious way. And a tremendous amount of new products will get into the market to meet the bandwidth demand in the next few years. But how to ensure a qualified product is not easy, especially the key interconnect components - optical transceivers. This article will focus on the challenges of the 400G transceiver test and the key items in the 400G optical module test.
Both 200G and 400G Ethernet have brought several new optical transceiver types, and the IEEE and MSA (multi-source agreements) group have defined the specific form factors: QSFP-DD, OSFP, and CFP8. The electrical interfaces of these transceivers use either 16 electrical lanes of 28Gb/s per lane signaling with NRZ (non-return to zero) modulation, or the newer 4- or 8-lanes of 56Gb/s per lane signaling with PAM4 (4-level pulse amplitude) modulation.
|Physical Medium Dependent||Host Electrical I/F||Mode||No. of Fibers||Reach Range||Encoding Method|
|400GBASE-SR16||16x 25Gb/s||Single mode||16||100 meters||NRZ|
|400GBASE-DR4||8x 50Gb/s||Single mode||4||500 meters||PAM4|
|Single mode||8 WDM||2 kilometers||PAM4|
|Single mode||8 WDM||10 kilometers||PAM4|
|200GBASE-DR4||8x 50Gb/s||Single mode||4||500 meters||PAM4|
|200GBASE-FR4||8x 50Gb/s||Single mode||4 WDM||2 kilometers||PAM4|
|200GBASE-LR4||8x 50Gb/s||Single mode||4 WDM||10 kilometers||PAM4|
Higher speeds and the utilization of PAM4 modulation do bring great improvements in throughput but also result in high complexity at the physical layer, and it causes signal transmission errors easily.
The first problem is, the higher lane speed in 400G electrical interfaces means more noise (also called signal-to-noise ratio) in signal transmission. And the high signal-to-noise ratio causes an increased bit error rate (BER), which in turn affects the signal quality.
Furthermore, on the physical appearance layer, for 400G optical modules, its high-speed interfaces include more electrical input interfaces, electrical output interfaces, optical input interfaces, optical output interfaces, and other power and low-speed management interfaces. All the performance of these interfaces should be made to a complaint of 400G standards. However, the size of the 400G transceivers is similar to the existing 100G transceivers, the integration of those interfaces needs more sophisticated manufacture technology, as well as corresponding performance tests to ensure the quality of those modules.
At the same time, the complex of the 400G transceiver test also brings new challenges for the optical module vendors. To ensure the transceiver quality for users, vendors have to attach great importance to the transceiver test equipment and R&D technical. How to make sure the new products supporting the 400G upgrade while dampening associated development and manufacturing test costs that can hamper competitive pricing models, is what they should deal with.
Though 400G Ethernet standards have been approved for years, the whole industry including OEMs and network/data center operators are still addressing basic connectivity issues, attempting to solve problems from transceiver reliability to link flap, from excessive frame error to excessive packet loss. For transceiver vendors, product quality testing is fundamental to build reliable connections with customers. Let's have a look at the several main test items in the 400G transceiver testing process.
ER (extinction ratio) is an important indicator to measure the performance of the 400G optical transmitters, and also the most difficult one. ER is the optical power logarithms ratio when the laser outputs the high level and low level after electric signals are modulated to optical signals. The ER test can show whether a laser works at the best bias point and within the optimal modulation efficiency range. OMA (outer optical modulation amplitude) can measure the power differences when the transceiver laser turns on and off, which tests the transceiver working performance in another aspect. Both the ER and the average power can be measured by mainstream optical oscilloscopes.
400G transceiver is a more complicated integration compared with the existing QSFP28 and QSFP+ modules, which also puts higher requirements for the test of its forwarding performance. RFC 2544 defines the following baseline performance test indicator for networks and devices: throughput, delay, and packet loss rate. In this test procedure, the electrical and optical interfaces will be tested and make sure the signal quality they transmitted and received will not get distortion.
Different from the single eye diagram of NRZ modulation in 100G optical transceivers, the PAM4 eye diagram has three eyes. And PAM4 doubles the bit bearing efficiency compared with NRZ, but it still has noise, linearity and sensitivity problems. IEEE proposes using PRBS13Q to test the PAM4 optical eye diagram. The main test indicators are eye height and width. By checking the eye height and width in the test result, users can tell the signal linearity quality of the 400G transceiver is good or not.
Jitter tests are mainly designed for output jitters of transmitters and jitter tolerance of receivers. The Jitter includes random Jitter and deterministic Jitter because deterministic jitter is predictable when compared to random jitter, you can design your transmitter and receiver to eliminate it. In a real test environment, the Jitter test is operated together with the eye diagram test to check the 400G transmitter and receiver performance.
In this testing procedure, 400G optical transceiver will be plugged into the 400G switches to test its working performance, BER and error tolerance ability in a real environment. As mentioned above, the increased BER in 400G optical transceiver lanes is higher because of the higher speed, which leads to transmission problems in most 400G links. Therefore, FEC (forward error correction) technology is applied to improve signal transmission quality.
FEC adds a pre-determined number of redundant bits into a data transmission that are error-checking bits (encoding these with the data). The error-checking bits are then used by the receiver of the data transmission to decode and correct errored bits. FEC provides a way to send and receive data in extremely noisy signaling environments, making an error-free data transmissions in 400G link as possible.
Therefore, in this real condition test process, the original BER of the optical transceiver and the corrected BER by FEC should be tested, verify whether the whole link performance is affected when a predetermined random error symbol or frequency deviation occurs.
Driven by 5G, artificial intelligence (AI), virtual reality (VR), Internet of Things (IoT), and autonomous vehicles, though there are multiple technical transceiver test issues needing to be resolved, the booming trend of the 400G Ethernet market cannot stop. Lots of manufacturers like Cisco, Arista, Fisinar, etc. and test solution providers like Keysight and Ixia, have promoted their own 400G product solutions to the market. Under this situation, for some smaller optical module vendors, the 400G transceiver test is one of the key points they should consider, because how to improve the quality of the 400G products and supply speed will determine how much profit they get from the 400G market. Know more about the 400G Ethernet Market Current and the Future to make preparation for the coming fast-speed era.