- Mia
- September 10, 2023
- 8:14 am
Harry Collins
Answered on 8:14 am
For 100G mode with QPSK modulation, the FEC type could be oFEC (OpenROADM) or SDFEC (Soft Decision FEC) with probabilistic constellation shaping.
For 200G mode with QPSK modulation, the FEC type could be SDFEC with probabilistic constellation shaping.
For 400G mode with 16-QAM modulation, the FEC type could be SDFEC with probabilistic constellation shaping.
The 400G-BIDI module can work in different ways depending on the switch and the application. It can use different types of FEC (Forward Error Correction) to improve the signal quality. FEC is a method of adding extra bits to the data to detect and correct errors.
One way is to use the 400G-SR4.2 or 4x 100G-SR1.2 mode, where the switch does the FEC and the module does not. The FEC used here is called KP-FEC, which is a type of Reed-Solomon FEC.
Another way is to use the 4x 100G-BIDI (100G-SRBD) mode, where the module does the FEC and the switch does not. The module uses a different FEC for each pair of fibers, which can work with the existing 100G-BIDI modules. This way, you can keep using your old 100G-BIDI modules with the new 400G-BIDI modules.
People Also Ask
AI Compute Clusters: Powering the Future
In recent years, the global rise of artificial intelligence (AI) has captured widespread attention across society. A common point of discussion surrounding AI is the concept of compute clusters—one of the three foundational pillars of AI, alongside algorithms and data. These compute clusters serve as the primary source of computational
Data Center Switches: Current Landscape and Future Trends
As artificial intelligence (AI) drives exponential growth in data volumes and model complexity, distributed computing leverages interconnected nodes to accelerate training processes. Data center switches play a pivotal role in ensuring timely message delivery across nodes, particularly in large-scale data centers where tail latency is critical for handling competitive workloads.
Comprehensive Guide to 100G BIDI QSFP28 Simplex LC SMF Transceivers
The demand for high-speed, cost-effective, and fiber-efficient optical transceivers has surged with the growth of data centers, telecommunications, and 5G networks. The 100G BIDI QSFP28 (Bidirectional Quad Small Form-Factor Pluggable 28) transceiver is a standout solution, enabling 100 Gigabit Ethernet (100GbE) over a single-mode fiber (SMF) with a simplex LC
NVIDIA SN5600: The Ultimate Ethernet Switch for AI and Cloud Data Centers
The NVIDIA SN5600 is a cutting-edge, high-performance Ethernet switch designed to meet the demanding needs of modern data centers, particularly those focused on artificial intelligence (AI), high-performance computing (HPC), and cloud-scale infrastructure. As part of NVIDIA’s Spectrum-4 series, the SN5600 delivers unparalleled throughput, low latency, and advanced networking features, making
How Ethernet Outpaces InfiniBand in AI Networking
Ethernet Challenges InfiniBand’s Dominance InfiniBand dominated high-performance networking in the early days of generative AI due to its superior speed and low latency. However, Ethernet has made significant strides, leveraging cost efficiency, scalability, and continuous technological advancements to close the gap with InfiniBand networking. Industry giants like Amazon, Google, Oracle,
Understanding NVIDIA’s Product Ecosystem and Naming Conventions
Compute Chips—V100, A100, H100, B200, etc. These terms are among the most commonly encountered in discussions about artificial intelligence. They refer to AI compute cards, specifically GPU models. NVIDIA releases a new GPU architecture every few years, each named after a renowned scientist. Cards based on a particular architecture typically
Related Articles
800G SR8 and 400G SR4 Optical Transceiver Modules Compatibility and Interconnection Test Report
Version Change Log Writer V0 Sample Test Cassie Test Purpose Test Objects：800G OSFP SR8/400G OSFP SR4/400G Q112 SR4. By conducting corresponding tests, the test parameters meet the relevant industry standards, and the test modules can be normally used for Nvidia (Mellanox) MQM9790 switch, Nvidia (Mellanox) ConnectX-7 network card and Nvidia (Mellanox) BlueField-3, laying a foundation for
AI Compute Clusters: Powering the Future
In recent years, the global rise of artificial intelligence (AI) has captured widespread attention across society. A common point of discussion surrounding AI is the concept of compute clusters—one of the three foundational pillars of AI, alongside algorithms and data. These compute clusters serve as the primary source of computational
Data Center Switches: Current Landscape and Future Trends
As artificial intelligence (AI) drives exponential growth in data volumes and model complexity, distributed computing leverages interconnected nodes to accelerate training processes. Data center switches play a pivotal role in ensuring timely message delivery across nodes, particularly in large-scale data centers where tail latency is critical for handling competitive workloads.
Comprehensive Guide to 100G BIDI QSFP28 Simplex LC SMF Transceivers
The demand for high-speed, cost-effective, and fiber-efficient optical transceivers has surged with the growth of data centers, telecommunications, and 5G networks. The 100G BIDI QSFP28 (Bidirectional Quad Small Form-Factor Pluggable 28) transceiver is a standout solution, enabling 100 Gigabit Ethernet (100GbE) over a single-mode fiber (SMF) with a simplex LC
NVIDIA SN5600: The Ultimate Ethernet Switch for AI and Cloud Data Centers
The NVIDIA SN5600 is a cutting-edge, high-performance Ethernet switch designed to meet the demanding needs of modern data centers, particularly those focused on artificial intelligence (AI), high-performance computing (HPC), and cloud-scale infrastructure. As part of NVIDIA’s Spectrum-4 series, the SN5600 delivers unparalleled throughput, low latency, and advanced networking features, making
How Ethernet Outpaces InfiniBand in AI Networking
Ethernet Challenges InfiniBand’s Dominance InfiniBand dominated high-performance networking in the early days of generative AI due to its superior speed and low latency. However, Ethernet has made significant strides, leveraging cost efficiency, scalability, and continuous technological advancements to close the gap with InfiniBand networking. Industry giants like Amazon, Google, Oracle,
Understanding NVIDIA’s Product Ecosystem and Naming Conventions
Compute Chips—V100, A100, H100, B200, etc. These terms are among the most commonly encountered in discussions about artificial intelligence. They refer to AI compute cards, specifically GPU models. NVIDIA releases a new GPU architecture every few years, each named after a renowned scientist. Cards based on a particular architecture typically