Thursday, March 28, 2024

Coherent reports major advance in VCSELs

Coherent reported a  significant breakthrough in improving the bandwidth of its vertical-cavity surface-emitting laser (VCSEL), paving the way for use in next-generation optical interconnects that will operate at 200 Gbps per lane and thereby enabling optical transceivers with 1.6 Tbps capacity.

This improvement comes in the form of lithographic-aperture VCSELs, marking a significant departure from the oxide-aperture VCSELs that are the state-of-the-art VCSELs used in AI networks today. 

Oxide-aperture VCSELs are currently powering optical transceivers operating at 100G per lane data rates, including 800G (8x100G) multimode optical transceivers, but have reached their bandwidth limit of approximately 27 GHz. 

The new lithographic-aperture VCSELs feature exceedingly small apertures with high reproducibility and reliability. With this innovative design, it is feasible to achieve a device bandwidth of approximately 45 GHz, which, when combined with recent advances in digital signal processing technology, can pave the way to 1.6T (8x200G) multimode optical transceivers.

Multimode transceivers and links are ideally suited for short-reach connectivity in data centers, spanning artificial intelligence (AI), high-performance computing (HPC), and networking applications.

A technical paper describing this leap in performance was presented by Dr. Stefano Tirelli,  Senior Manager, VCSEL Technology, at OFC 2024.

“As a leading supplier of VCSELs for multimode 800G optical transceivers, we are pleased to extend our technological leadership on our roadmap toward next-generation 200G VCSELs,” said Chair and CEO Dr. Vincent D. (“Chuck”) Mattera, Jr. “The development of a reliable, lithographic-aperture VCSEL is a step in the right direction. While we have more work to do, we are very pleased with the progress we have made so far.”

Keysight bids for Spirent, topping Viavi’s offer

 Keysight Technologies announced its intention to acquire the entire issued and to be issued share capital of Spirent Communications PLC for a cash consideration of 199 pence per Spirent share, which values Spirent at approximately £1,158 million (US$ $1,463 million1), on a fully diluted basis. Spirent shareholders will also be entitled to receive a special dividend of 2.5 pence per Spirent share, in lieu of any final dividend for the year ended 31 December 2023 (together with the cash consideration of 199 pence per share, the “Offer”).

Spirent’s board of directors intend to unanimously recommend the Offer which is conditional on, among other things, the approval of the requisite majority of Spirent Shareholders at the Court Meeting and the Spirent General Meeting, as well as regulatory approval as laid out in the Offer announcement.

Keysight said it views Spirent’s business as highly attractive and aligned with Keysight’s long-term software-centric solutions strategy. Keysight believes there is strategic synergy between the two businesses, and that the parties’ respective complementary products will help customers address their complex engineering challenges. 

Commenting on the acquisition, Satish Dhanasekaran, President & Chief Executive Officer of Keysight, said: “Spirent has a differentiated portfolio, which is a strong fit for Keysight. Both companies share a common focus on empowering and accelerating high-value solutions for customers. In Keysight, Spirent will join a platform of significant customer scale with the capacity to provide the necessary capital and resources to help Spirent grow and accelerate the delivery of its strategic vision. Keysight’s long-term customer relationships, industry expertise and global reach will help Spirent drive product development and execute on its full potential. With similar cultures valuing customer-centricity and high-performance, we believe that Keysight will be an excellent home for Spirent to thrive and deliver sustainable, long-term growth. Our superior Offer recognizes the value of Spirent’s achievements to-date, and the exciting prospects of the combination of our complementary product portfolios to provide end-to-end solutions for customers across their lifecycle needs.”

Viavi to acquire Spirent, consolidating test & measurement companies

VIAVI Solutions agreed to acquire Spirent Communications, a global provider of automated test and assurance solutions for networks, cybersecurity, and positioning, for approximately £1,005 million or $1,277 million, based on the exchange rate on March 4, 2024.In addition, Silver Lake is making a US$400 million long-term strategic investment in Viavi in connection with the acquisition. Viavi said it believes that acquiring Spirent will significantly...

Lumentum enhances its 800ZR+

Lumentum announced significant performance enhancements to its 800ZR+ transceivers. These improvements enable the 800ZR+ modules to serve a wider range of applications, including expanded use in metro and regional networks through the ability to interface directly into routers. Additionally, the enhanced performance modes now make 400ZR+ and 600ZR+ operation suitable for true long-haul applications, with OSNR and dispersion tolerance enabling reaches exceeding 2000 km at 400 Gbps. 

Lumentum has optimized its latest generation of 800ZR+ modules to maximize performance, while maintaining a compact QSFP-DD and OSFP form factor. These modules employ Lumentum’s latest hybrid photonic integrated circuit technology, which incorporates both the company’s leading edge indium phosphide photonic integrated circuits and silicon photonics. The performance and use case flexibility provided by these modules mark a significant step in the evolution of high-speed communications networks.

“Lumentum's unique photonic component capabilities and innovative module designs, together with our partners' latest generation coherent DSPs, enable us to provide solutions that dramatically increase scalability for both cloud and telecom network operators,” said Wupen Yuen, President of Cloud and Networking at Lumentum. “Our advancements also help address the challenges of scaling data center interconnects for the bandwidth needs related to artificial intelligence and machine learning."

These modules are currently available for sampling.

Credo and Wistron demo 800G Linear Receive Optics

Credo hosted a demonstration of Linear Receive Optics (LRO), using its Credo Dove 850 DSP, along with the 51.2 Tbps Wistron DS-6183-64O 64-port 800GbE switch. 

In an LRO transceiver, or Active Optical Cable (AOC), only the transmit path from the electrical input to the optical line side output includes a DSP for signal retiming and equalization. 

All three 800G OSFP LRO modules provided by Credo partners for this demonstration seamlessly “plug and play” with the 51.2T Wistron switch, a tremendous benefit in mass deployment.

The Dove 850 is a unidirectional 8 x 112 Gb/s DSP purpose built for the LRO architecture. The Dove 850 DSP aims to address the inherent weakness of a Linear Pluggable Optics (LPO) implementation by facilitating IEEE compliant optical transmit signaling and easing the deployment burden on system operators by eliminating manual, per port tuning. The result is lower bit error rates, enhanced sensitivity, reduced performance variation, and improved resilience to different switch ASICs, PCB traces, optical components, and environmental conditions. Dove 850 allows transceiver vendors to optimize module cost by employing a variety of optics, including VCSELs, silicon photonics, EML or thin-film lithium niobate as dictated by the application requirements.

Wistron’s DS-6183-64O is a compact 2RU 64 port 800G OSFP Ethernet Switch, built with Broadcom Tomahawk 5 chip, providing high bandwidth, high throughput, and reliable solutions. Support up to 102.4Tbps of switching I/O bandwidth and non-blocking switching fabric, delivering wire-speed L2 and L3 packet forwarding capabilities on all ports for Low-latency data transmission and support for state-of-the-art instrumentation feature. With the advanced system integration technique, all OSFP ports on DS-6183-64O support up to 800GbE and can be configured as 2 x 400GbE and also support 18W output. For long-range DCI application, 8 OSFP ports of DS-6183-64O could even support 25W in each port. DS-6183-64O is loaded with ONIE (Open Network Install Environment), which supports the installation of compatible NOS (Network Operating System) software, like SONiC.

“Linear Receive Optics are pushing the industry forward. Credo is streamlining implementation for module vendors while addressing the need for improved energy efficiency across the data center,” said Anderson Chiu, General Manager of Networking Development, Enterprise and Networking Business Group of Wistron. 

Coherent samples bi-di circulator adaptor

Coherent introduced a bi-di circulator adaptor that enables bi-directional transmission over a single fiber between two transceivers. By enabling bi-directional transmission, the fiber count is reduced by a factor of two, which allows significant cost and space savings.

The Coherent bi-di circulator adaptor is a completely passive device, compatible with all transceiver LC connector ports, offering low insertion loss as well as high directivity in a compact footprint. It is available in O-band, C-band, and L-band versions, and customized designs can be considered as well.

“The urgent demand for improved utilization of the fiber plant with markets as diverse as AI/ML applications within the data center and the broadband access market shows the real need for simple, cost-effective solutions like our bi-di adaptor. By combining our fully vertically integrated supply chain, including the core garnet chips, with our fully automated assembly process, we can deliver a high-volume solution that brings significant value to our customers,” said Dr. Beck Mason, Executive Vice President, Telecommunications.

Dell'Oro: Telecom Capex Declined in 2023


Preliminary findings show that worldwide telecom capex, the sum of wireless and wireline/other telecom carrier investments, declined for the full year 2023 in nominal USD terms, recording the first contraction since 2017, according to a recent report from Dell'Oro Group. This deceleration in the broader capex spend is consistent with the aggregate telco equipment slump previously communicated for the six Dell'Oro telecom programs (Broadband Access, Microwave Transmission & Mobile Backhaul, Optical Transport, Mobile Core Network, Radio Access Network, Service Provider Routers & Switch).

"The fundamental challenges have not changed. Operators have a fixed capital intensity budget and capex is largely constrained by the revenue trajectory," said Stefan Pongratz, Vice President and analyst with Dell'Oro Group. "What is complicating the situation is that the revenue pie remains fixed. Following some positive developments amidst the peak of the Covid-19 pandemic, our analysis shows that operator revenue growth slowed in 2023 and has more or less remained stagnant over the past decade. And based on the guidance, operators, in general, are not overly optimistic that emerging opportunities with generative AI, edge computing, enterprise 5G, FWA, and 5G-Advanced will expand the pie," continued Pongratz.

Additional highlights from the March 2024 Telecom Capex report:

  • Global carrier revenues are expected to increase at a 1 percent CAGR over the next 3 years.
  • Market conditions are expected to remain challenging in 2024. Worldwide telecom capex is now projected to decline at a mid-single-digit rate in 2024 and at a negative 2 to 3 percent CAGR by 2026.
  • The mix between wireless and wireline remains largely unchanged, reflecting challenging times still ahead for wireless. Wireless related capex is on track to decline at a double-digit rate in the US in 2024.
  • 5G era capital intensity ratios peaked in 2022 and are on track to approach 15 percent by 2026, down from 18 percent in 2022.

TRUMPF demos 100G VCSEL for 800G transceiver

 TRUMPF Photonic Components demonstrated 100Gbps VCSEL performance in an 800G tranceiver by Optomind. 

TRUMPF argues that for interconnect distances of up to 100 meters, VCSELs are the best solution in terms of power, cost and the productivity of SR modules. TRUMPF is offering both VCSELs and photodiodes as a matching pair solution, in singlets, 1x4 arrays and 1x12 arrays for 14G and 25G for the NRZ applications. The same is offered for 56G PAM4 applications. The VCSELs are specifically designed and implemented to meet the requirements of hyperscale data centers including for AI/ML applications, high-performance computing systems, and other bandwidth-intensive applications, as they deliver high performance and reliable data transmission at high speeds.

“We are delighted to have achieved a performance of 800Gbps in our transceiver leveraging innovative optics technology and TRUMPF’s VCSEL” commented Yung Son, Chief Marketing Officer of Optomind. “We look forward to solidifying our partnership with TRUMPF as the strategic best-in-class VCSEL supplier for distinguished transceiver and active optical cable (AOC) to our customers,” he added.

“We are pleased to have collaborated with Optomind to demonstrate the performance of our VCSEL at PAM4 112Gbps/ch in their transceiver, which validates the use of it in a real-world application,” says Ralph Gudde, Vice President of Marketing and Sales at TRUMPF Photonic Components. “A full-featured version of our 100G VCSEL is planned for production release this summer. TRUMPF is diversified, has a strong technology background and brings solid long-term commitment as a technology partner and a key supplier into the datacom business.” he explained.

MaxLinear announces “Rushmore” 200G/lane PAM4 DSPs

Building on its Keystone family of 5nm 100G/lane DSPs, MaxLinear announced the development of its fourth generation 200G/lane PAM4 SERDES (Serializer/Deserializer) and DSP family, Rushmore. 

The Rushmore family encompasses a diverse array of variants tailored to meet the evolving needs of modern data centers and telecommunications networks. This includes support for fully re-timed optical modules, half re-timed optics, and active electrical cables (AECs). Moreover, MaxLinear is introducing both 8x200G-to-8x200G re-timers and 8x100G-to-4x200G gearbox variants, providing customers with a comprehensive suite of solutions to address their specific requirements.

The devices will be produced by Samsung Foundry using its leading-edge-CMOS-based process.

“The AI/ML revolution has accelerated the pace of innovation across the industry, and that is also the case at MaxLinear,” said Drew Guckenberger, Vice President of High Speed Interconnect at MaxLinear. “Starting from our highly successful Keystone 5nm 100G/lane DSP family, the amount of additional innovation that has gone into the new Rushmore 200G/lane DSP family is incredible. The impacts on power and performance our partners will see are expected to be significant for the industry as a whole as we scale into the new AI/ML world.”

OE Solutions extends its transceiver portfolio

OE Solutions introduced two new Digital Coherent Optical (DCO) transceiver modules for applications in access networks. Both are available with optional EDFA booster amplifier and tunable optical filter.

  • Reduced-power 100G QSFP28 DCO transceiver -- a result of OE Solutions’ continued focus on reducing the power consumption of coherent transceivers for access applications. Current improvements to the designs enable a device with a maximum power consumption below 10W allowing the transceivers to be used in more network elements.  The 100G QSFP28 DCO without booster amplifier exhibits a power consumption of less than 10W. It will sample in May of this year and be generally available in June. 
  • 400ZR QSFP-DD DCO -- available with or without EDFA booster amplifier and tunable optical filter, is the first product in a series of 400G DCO transceivers that will include also ZR+ and OpenROADM variants in the future. The 400ZR device leverages 60GB components and 16-QAM encoding. Samples of the 400ZR QSFP-DD DCO transceiver will be available in August, and generally availability will be in October of this year.

“We are excited about this enhancement of our DCO portfolio giving our customer additional options for optimizing access networks to accommodate the growing bandwidth needs at the edge of the network”, said Walter Lee, Vice President and General Manager of the Optical Transceiver Business Unit. He continued, “We will continue to focus our development on coherent transceiver variants targeting the particular needs of access networks with their broad range of requirements and deployment environments”.

MACOM demos 100G/Lane Switch to Server Link with LPO

MACOM is demonstrating switch-to-server interoperability of its linear drive 100G per lane components:

  • MACOM PURE DRIVE Transimpedance Amplifiers and Laser Drivers
  • Broadcom’s BCM57608 Series Ethernet Network Interface Card (“NIC”)
  • Broadcom’s Tomahawk 5 Ethernet Switch Chip

The switch-to-server link represents an important use case for the linear drive architecture, which can provide an optimized interconnect solution for Cloud infrastructure. Linear drive modules eliminate the Digital Signal Processor (DSP) from pluggable optical modules enabling lower power consumption, improving signal latency and reducing cost.

“Due to linear drive’s benefits, the demand for linear pluggable optical (LPO) modules with MACOM PURE DRIVE technology is expanding. We are pleased to demonstrate a working example of 100G/lane switch-to-server connectivity leveraging this technology,” said Ryan Latchman, Vice President, High-Performance Connectivity Product Marketing, MACOM.

Wednesday, March 27, 2024

Celestial AI raises $175M series-C for its Photonic Fabric

Celestial AI, a start-up developing an optical interconnect for disaggregated, exascale compute and memory clusters, raised $175 million in its Series C funding.

Celestial AI is working on Optical Compute Interconnect (OCI) to enable disaggregation of scalable data center memory and accelerated computing. The company says its Photonic Fabric is unconstrained by package beachfront limitations and can deliver data to any location on the compute die, directly to the point of consumption. It promises greater scalability and more than 10X lower latency and power consumption compared with any optical interconnect alternative, such as Co-Packaged Optics (CPO).

The latest funding round wass led by Thomas Tull’s U.S. Innovative Technology Fund (USIT). In addition to USIT, the round included participation from new and existing investors including AMD Ventures, Koch Disruptive Technologies (KDT), Temasek, Temasek’s wholly owned subsidiary Xora Innovation, IAG Capital Partners, Samsung Catalyst, Smart Global Holdings (SGH), Porsche Automobil Holding SE, Engine Ventures, M-Ventures and Tyche Partners.

“This highly oversubscribed financing round is another testament to Celestial AI’s position to disrupt the market, and reflects the magnitude of business opportunity before us,” said Dave Lazovsky, Celestial AI founder and CEO. “Today’s advanced AI models require exponentially more I/O bandwidth and memory capacity but are currently constrained by low bandwidth and high latency. Our Photonic Fabric™ technology is rapidly becoming the de facto standard for Accelerated Computing optical interconnectivity, driven by unparalleled performance and efficiency.”

Marvell launches 1.6T optical DSP, 3D SiPho Engine

At #OFC24, Marvell outlined its broad-ranging strategy to optimize optical technology to meet the rising performance and power requirements of AI and cloud data centers.

Products and use cases being unveiled at OFC include:

  • Nova 2: The industry's first 1.6T optical DSP featuring 200G per lane electrical and optical I/O, Marvell Nova 2 will enable module integrators to produce industry-standard optical transceivers delivering the bandwidth density required by switches, AI accelerators and other devices. 200 Gbps per lane I/O is a defining feature of next-generation data center architectures.

Nova 2 is sampling to customers now. Nova, a 1.6T optical DSP featuring 100G per lane electrical I/O and 200G per lane optical interfaces, is now generally available.

  • Marvell 3D Silicon Photonics Engine: The Marvell 3D SiPho Engine combines hundreds of components needed for optical communications into a single device. Featuring 200G I/O, 3D SiPho Engine delivers 2x the bandwidth, 2x the input/output (I/O) bandwidth density, and 30% lower power per bit versus comparable devices with 100 Gbps electrical and optical interfaces1. Potential applications include modules for AI clusters.

  • Spica Gen2-T: A new category of optical DSPs, Spica Gen2-T is designed for transmit retimed optical (TRO) modules, an emerging class of modules for AI and cloud use cases. Transmit-only DSPs enable TROs to reduce power consumption by more than 40% compared to conventional transceivers2. Spica Gen2-T is the fourth 800 Gbps optical DSP from Marvell and part of its effort to optimize optical DSPs, along with other cloud and AI silicon, for specific use cases and applications.

  • COLORZ 800: First Demo of Probabilistic Constellation Shaping at 800 Gbps and Industry Momentum. Probabilistic Constellation Shaping (PCS) effectively doubles the reach of coherent DSP modules employed for long-distance communications. Marvell will showcase COLORZ 800 modules linked to a Teralynx 10 switch running in PCS mode capable of providing full capacity for up to 1000 km.

Marvell COLORZ 800 technology will also be on display in booths from Arista Networks (#4901) and the Optical Internetworking Forum (OIF) multi-vendor interoperability demonstration (#1323).

  • O-band Optics: Xi Wang, Ph.D., vice president of marketing for Optical Connectivity at Marvell, examines the growing use cases for coherent DSPs leveraging the O-band optics in data centers. At higher bandwidths, O-band optics can potentially cover greater lengths than PAM4 DSPs while consuming less power than today's coherent DSPs. The Market Watch panel takes place on Tuesday, March 26 at 12:30-2:00 p.m. in Theater 1.

  • Juniper Networks, Coherent Corp., and Marvell Collaborate on ZR/ZR+ Transport Solution: The three companies will demonstrate a combined solution featuring Marvell Orion, the industry's first 800 Gbps coherent DSP for long-distance data center interconnects and telecommunications networks, Coherent's module technology and Juniper's routing platform.

  • Customer Momentum for Orion with InnoLight and HGGenuine: Module manufacturing leaders will highlight new 800G ZR/ZR+ modules powered by Orion for data center interconnects, IP over DWDM, and other use cases.

"Over the last two decades, optical data rates have increased 1000x while energy per bit has declined 100x3, and in the process transformed data infrastructure and enabled the rise of services like streaming media, cloud computing, smartphones and AI," said Dr. Radha Nagarajan, senior vice president and chief technology officer, Optical Platforms at Marvell and the 2024 recipient of the David Richardson Medal for significant contributions to optical engineering. "The industry is now entering a new chapter and Marvell is looking forward to collaborating with its partners in extending the horizons of this far-reaching technology."


Lumentum launches 200G per Lane InP Components, 800ZR+

 Here are highlights of Lumentum's demonstrations at #OFC24:

  • 200G per Lane InP Components: Lumentum extends its indium phosphide (InP) transceiver component leadership with its new 200G Lens Integrated Photodiode (LIPD) that supports operation up to 200 Gbps per lane, enabling next-generation 800G and 1.6T optical transceivers for large-scale AI and ML applications. Featuring high responsivity and bandwidth, the 200G LIPD is designed for seamless integration with flip-chip bonding techniques and is now available for sampling. The 200G LIPD complements Lumentum’s high-volume, production ready 200G externally modulated lasers (EMLs), giving optical transceiver manufacturers best-in-class transmit and receive components for their next-generation 800G and 1.6T solutions.
  • Ultra-High Power 400 mW 1310 nm DFB Laser: Marking a milestone in semiconductor laser technology, Lumentum introduces an ultra-high output power 1310 nm DFB laser capable of delivering 400 mW of optical power at 50°C. This beta release is designed for co-packaged optics and silicon photonic 800G and 1.6T transceiver applications, offering a solution to reduce the required number of lasers, boosting power efficiency and reliability in large scale AI and ML infrastructure. Lumentum will demonstrate its ultra-high power 1310 nm DFB laser at its booth.
  • Data Center Connectivity Demo including Linear Receiver AOC: Lumentum will demonstrate data center connectivity scenarios with a 51T switch, a server with a 4x100G NIC, and several of its Cloud Light™ 800G optical modules operating in a loopback configuration. The performance of a linear receive-only AOC connecting an 800G switch port and the server NIC will be shown in the demonstration.
  • 400G QSFP112 DR1.2: Lumentum will also demonstrate its Cloud Light 400G QSFP112 DR1.2 module, which features a 4x100G electrical interface and produces two wavelengths (1290 nm and 1310 nm), each operating at 200G.
  • 400/800ZR+ Modules: Lumentum will demonstrate its 800ZR+ OSFP module operating at 118 GBaud 16 QAM. Designed to be used in direct router interface applications, performance will be demonstrated over a more than 400 km amplified optical link, highlighting the suitability for extended DCI and metro applications. In addition, a high power 400ZR+ QSFP-DD will be operated over the same link, showing the high performance possible in current long-haul networks using pluggable modules.
  • OIF 400ZR/800ZR Interoperability and CMIS: At OIF's booth #1323, Lumentum highlighted the interoperability of its 800ZR and 400ZR transceivers with QSFP-DD and OSFP equipped switches and routers for IP over DWDM, supporting up to 800 Gbps and complying with OIF standards. Additionally, demonstrations of Lumentum's OpenZR+ and 400ZR QSFP-DD transceivers will emphasize the benefits of the CMIS standard for efficient module management and interoperability, enabling precise control over status, optical channels, and power across various applications.
  • D22 Series Pump Module: This state-of-the-art, dual-chip 980 nm pump laser module offers individual control over each emitter. It integrates Lumentum’s highly reliable and efficient 980 nm diode lasers into a compact, hermetically sealed design, delivering exceptional optical power density. The D22 series provides a noise-free, narrowband spectrum and maintains peak performance across various temperatures and current levels. It is ideally suited for applications demanding superior spectrum control and high-power output, offering up to 2x1100 mW.
  • 49M Series Multimode Pump Module: Available with 4 or 6 emitters at 980 nm, the 49M series delivers powerful output of 40 W or 60 W while maintaining a compact footprint. Its innovative design simplifies thermal management, supporting both air- and water-cooling systems. This versatility makes the 49M Series perfect for a range of applications, including vacuum, space, and terrestrial EDFA pumping.
  • Tunable SFP28 ER Module: Designed for industrial temperatures, this high-performance optical transceiver supports 25 Gbps DWDM links over distances up to 40 km. It offers a compelling upgrade for cable MSOs facing increasing bandwidth demands driven by DOCSIS 4.0 and FTTH deployments. Additionally, it enhances 5G transport network connectivity, leveraging existing C-band fiber infrastructure. Lumentum demonstrated the capabilities of this new Tunable SFP28 ER module at its booth.

"We are excited to share with the industry our latest innovations, including our highly differentiated and broad portfolio of photonics products and technology that enable cloud operators and cloud infrastructure providers to further scale AI data centers,” stated Lumentum President and CEO Alan Lowe. "Our showcase at OFC 2024 highlights Lumentum's commitment to meeting the ever-increasing demands of AI, machine learning (ML), and next-generation communications networks.”


Semtech demos Linear Pluggable Optics

At #OFC24, Semtech hosted the following live demonstrations:

  • 400Gbps SR4 LPO modules and 800G SR8 LPO modules, featuring Semtech’s DirectEdge GN1858 quad VCSEL driver and GN1856 quad TIA, interoperating in isolation with latest switch ASIC technology as well as with the latest Network Interface Card (NIC) technology, emulating a Top of Rack to NIC arrangement
  • 400Gbps DR4 LPO module featuring Semtech’s FiberEdge MZM driver and GN1814 quad TIA, interoperating with latest switch ASIC technology
  • 400Gbps FR4 LPO module featuring Semtech’s FiberEdge EML driver and GN1817 quad TIA, interoperating with latest switch ASIC technology
  • 800Gbps DR8 LPO module featuring Semtech’s FiberEdge EML driver, GN1825 octal TIA and GN8089 Semtech’s own PIC (Photonic Integrated Circuit), interoperating with latest switch ASIC technology

In addition, 400Gbps SR4 and DR4 LPO modules with Semtech’s PMDs will be interoperating with latest switch ASIC technology and other LPO modules from multi-vendor ecosystem in OIF’s booth #1323.

“The demonstration of a vast number of ready-to-adopt multi-mode and single-mode LPO modules in Semtech’s booth at OFC this year to the extensive PMDs already present in the FiberEdge product line, reflects Semtech’s commitment to support LPO and take it to the next adoption level,” said Nicola Bramante, senior product line manager for Semtech’s Signal Integrity Products Group. “Reliable signal integrity, equalization and flexibility are key to make LPO successful, and Semtech’s DirectEdge and FiberEdge PMDs are the perfect candidates for this to match the needs of AI and ML high speed interconnects while reducing optical module power consumption by up to 50% versus DSP based solutions. Data centers are expected to utilize this power savings to deploy increased compute capability. LPO solutions also offer reduced latency and lower costs than DSP based solutions delivering additional value propositions for data centers.”


BT tests Adtran's 100ZR pluggable

 BT Group has conducted a field trial of Adtran’s Coherent 100ZR pluggable transceiver. The trial with BT Group was conducted across a complex network infrastructure, spanning 90km over a third-party optical line system, incorporating four ROADM nodes and six amplifiers. Utilizing a tight 50GHz ROADM channel, the trial achieved enhanced spectral efficiency. Operating at less than 5 watts, the trial also set new benchmarks for energy efficiency.

The Adtran Coherent 100ZR transceiver can be plugged directly into an operator‘s transmission unit. Featuring a purpose-built digital signal processor (DSP) co-developed with Coherent Corp that is optimized for data rates of 100Gbit/s and designed for easy integration with deployed infrastructure, the QSFP28-compatible transceiver facilitates the adoption of 100ZR coherent technology at the network edge.

“Adtran’s plug-and-play solution has enabled us to demonstrate the feasibility of boosting capabilities within existing networks and doing so in a cost-effective and scalable manner,” said Andrew Lord, senior manager of optics research at BT. “As we continue to innovate and push boundaries, technologies like this will be instrumental in achieving our goals for a greener, more efficient telecommunications landscape.”

“Together with BT Group, we’ve showcased a method for transforming network infrastructure with highly efficient, low-power solutions. It’s a strategy that requires significantly less equipment and fewer sites. The demo also proves the advantages of deploying one amplifier in the optical line system rather than multiple amplifiers within each transceiver module. This streamlined approach helps boost network performance, reduces operational spending and shrinks carbon footprint. As well as big potential savings – as much as 50% in some cases – this strategy offers a clear path to more sustainable network development,” commented Ryan Schmidt, GM of optical transport at Adtran. 


AST Mobile Tapes Out Space-to-Cellular ASIC

AST SpaceMobile, which is building the first and only space-based cellular broadband network accessible directly by everyday smartphones, has begun the tape-out phase for its Application-Specific Integrated Circuit (ASIC), in collaboration with TSMC.

AST SpaceMobile's BlueBird Block 2 program, the AST5000 ASIC is a novel, custom and low-power architecture developed to enable up to a tenfold improvement in processing bandwidth on each satellite, unlocking opportunities for seamless space-based cellular broadband services worldwide. 

The company says this  significant milestone marks the culmination of over four years of dedicated research, development, and engineering expertise, equivalent to an estimated 150 man-years of intensive work, as well as approximately $45 million of development.

"The commencement of the tape-out process for our ASIC, in collaboration with TSMC and other leaders in the semiconductor industry, is a testament to the unwavering commitment of our talented team," said Abel Avellan, Chairman and CEO of AST SpaceMobile. “As we advance towards the orbital launch of our Block 1 BlueBird satellites in the upcoming months, we are simultaneously making steady progress on the technology underlying our Block 2 satellites. This achievement underscores our relentless pursuit of innovation to address the growing demand for global connectivity, With the ASIC at the heart of our Block 2 program, we are a big step closer to ushering in a new era of satellite communications, helping to empower people with reliable connectivity, regardless of their location.”

AST SpaceMobile has achieved 3 Bits/Hertz using BlueWalker 3 and has planned capacity of up to 40 MHz per beam on its commercial satellites, which is planned to support 120 Mbps peak data rates. With our planned ASICs and 2,400 sq ft BlueBird satellite size, we expect to support up to 10,000 MHz of processing bandwidth per satellite in the future to enable the first and only space-based cellular broadband network.

AST SpaceMobile has more than 3,100 patent and patent-pending claims for its technology and operates state-of-the-art, vertically integrated manufacturing and testing facilities in Midland, Texas, which collectively span 185,000 square feet. The company recently secured strategic investment from AT&T, Google and Vodafone, as well as a new contract award with the United States Government through a prime contractor. The company has agreements and understandings with more than 40 mobile network operators globally, which have over two billion existing subscribers total, including Vodafone Group, Rakuten Mobile, AT&T, Bell Canada, Orange, Telefonica, TIM, Saudi Telecom Company, MTN, Zain KSA, Etisalat, Indosat Ooredoo Hutchison, Telkomsel, Smart Communications, Globe Telecom, Millicom, Smartfren, Telecom Argentina, Telstra, Africell, Liberty Latin America and others. Rakuten, American Tower and Bell Canada are also existing investors in AST SpaceMobile.


MaxLinear, Jabil, Intel collaborate on 800G silicon photonics

MaxLinear and Jabil announced the production availability of a family of 800G silicon photonics-based optical transceiver modules.

The new modules leverage Jabil’s recently-announced collaboration with Intel and their silicon photonics platform, renowned for its manufacturing efficiency and high reliability. Based on the high-volume, high-reliability platform from Intel and MaxLinear, Jabil will supply 800G-DR8, 2x400G-FR4 and 2x400G-LR4 optical modules to the market.

At the core of the new 800G module lies MaxLinear's 5nm 800G PAM4 DSP.

Intel’s volume-proven silicon photonics platform leverages Intel’s high-volume silicon manufacturing technologies, and includes on-chip laser sources fabricated, tested, and burned-in at wafer-scale to provide unsurpassed reliability at component level, as well as simplicity and efficiency in module integration.

“Our collaboration with Jabil brings a completely unique supply chain to the exploding AI/ML ecosystem,” said Drew Guckenberger, VP of High Speed Interconnect at MaxLinear. “The unique supply chains and geodiversity that Jabil and Intel bring are incredibly important for successfully scaling up networks in an accelerating demand environment like we are currently in.”

 “Jabil is excited to enter the optical module space as a direct supplier to many of our existing customers,” said Matt Crowley, SVP of Cloud and Enterprise Infrastructure at Jabil. “We are committed to this space long-term and to providing our customers with a unique value proposition they just cannot get anywhere else."


Dell'Oro: Campus switch backlog normalizes

In 4Q 2023 worldwide Campus Switch sales declined by 4 percent year-over-year (Y/Y), marking the first quarter of decline since 4Q 2020 and signaling the onset of a multi-quarter digestion cycle in the market, according to a new report from Dell'Oro Group.

Vendor performance was uneven during the quarter. Among the top five vendors, Cisco declined the most and was responsible for the bulk of the market decline. Huawei and Juniper declined modestly, while HPE and H3C grew by double digits.

“The revenue decline is mostly reflective of a digestion cycle that seems to be affecting vendors, products, and speeds to varying degrees and on different timelines,” said Sameh Boujelbene, Vice President at Dell’Oro Group. “Put simply, customers that received their orders early and vendors that were faster in fulfilling their backlogs will be the first to enter the digestion cycle. For that reason, vendor performance, regional performance, and even port speed performance was uneven during the quarter,” added Boujelbene.

Additional highlights from the 4Q 2023 Ethernet Switch – Campus Report:

  • Regionally, we observed a reversal of the trend seen over the last year. Regions such as North America (NA), Europe, Middle East and Africa (EMEA), and Asia Pacific (APAC) Excluding China, grew robustly in the prior quarters but saw a significant deceleration and, in some cases, a decline, in 4Q 2023. Meanwhile, China, which experienced sluggish sales in the past few quarters, showed some signs of recovery in the quarter.
  • 5/5.0 Gbps ports were up more than 60 percent, marking the seventh consecutive quarter of strong double-digit growth. This ongoing growth underscores significant improvements in supply, coupled with rising demand stimulated by the increasing adoption of Wi-Fi 6E Access Points (APs).

Tuesday, March 26, 2024

Lumentum demos 200G per Lane InP Components

Lumentum is extending its indium phosphide (InP) transceiver component leadership with its new 200G Lens Integrated Photodiode (LIPD) that supports operation up to 200 Gbps per lane, enabling next-generation 800G and 1.6T optical transceivers for large-scale AI and ML applications. Featuring high responsivity and bandwidth, the 200G LIPD is designed for seamless integration with flip-chip bonding techniques and is now available for sampling. 

The 200G LIPD complements Lumentum’s high-volume, production ready 200G externally modulated lasers (EMLs).

 Lumentum says the 200G Lens Integrated Photodiode could be used by optical transceiver manufacturers for best-in-class transmit and receive components for their next-generation 800G and 1.6T solutions.

Adtran shows ultra-low-power 100G PAM4 single-mode VCSEL

Adtran teamed up with Vertilas to develop a 100Gbit/s PAM4 single-mode vertical-cavity surface-emitting laser (VCSEL) with capabilities up to 1.6Tbit/s. 

The new VCSEL technology leverages a short-cavity single-mode design and is based on indium phosphide semiconductor material. This enables it to operate across both O-band and C-band wavelengths, delivering 100Gbit/s per channel. By stacking multiple single-mode lasers into arrays, this approach achieves throughput of 800Gbit/s and 1.6Tbit/s, ensuring superior performance with a significant reduction in power consumption and costs. This enhanced energy efficiency presents a major advantage for IP router and Ethernet switch faceplate density. 

Adtrans says the technology could be extended to support other applications from DR4/8/16 to various FR4 links within optical engines, all while maintaining power consumption below 200mW per lane.

The companies cite up to 80% reduction in power on the transmit optics compared to conventional solutions, with the new module consuming less than 2pJ/bit, including the laser driver. This significantly undercuts current industry standards, including those cited by co-packaged optics. 

“As the importance of data volume and speed escalates, the demand for increased bandwidth and enhanced energy efficiency accelerates. Our introduction of the market’s first 100Gbit/s PAM4 single-mode VCSEL technology, scalable to 1.6Tbit/s, epitomizes our proactive approach to tackling these emerging requirements. By pushing the limits of speed and significantly reducing power consumption, we’re creating tomorrow’s benchmarks,” said Ross Saunders, GM of optical engines at Adtran. “Our new technology promises to significantly improve intra-data center connectivity and AI application efficiencies, merging high performance with long-term sustainability. As we prepare for the leap towards 800Gbit/s and 1.6Tbit/s applications, we’re enhancing the foundations of efficient, high-capacity data ecosystems worldwide.”

"As applications such as generative AI proliferate and enterprises expand their AI clusters, the need for advanced optical interconnects capable of supporting Terabit connectivity has reached a critical juncture. Our 100Gbit/s PAM4 single-mode VCSEL technology meets this challenge head-on. It enables energy efficiency and the capacity for higher-density deployments while at the same time addressing manufacturing, testing and assembly costs,” commented Christoph Glingener, CTO of Adtran. “This technology supports throughput of up to 800Gbit/s and 1.6Tbit/s, significantly reducing operational expenses while enhancing network capabilities. Our approach underscores a commitment to delivering sustainable, high-capacity networking solutions that break through existing barriers, readying networks for the next wave of AI infrastructure development.”


AOI releases 400G, 800G Active Optical Cables based on Credo

Applied Optoelectronics, Inc (AOI) released two new multimode designs including a 400G QSFP-DD and an 800G OSFP. Both designs offers lower power consumption compared to previous generation solutions. 

  • The 400G design is available as either an Active Optical Cable (AOC) or 400GBASE-SR8 transceiver based on the Credo Seagull 452, an 8 x 50Gbps PAM4 DSP with integrated VCSEL drivers. 
  • The 800G design is available as either an Active Optical Cable (AOC) or 800GBASE-SR8 transceiver based on the Credo Dove 800, an 8 x 100Gbps PAM4 DSP. 

“AOCs and multimode transceivers are an integral part of AI and Ethernet networks, as an interconnect between the GPU server, and leaf and spine switches,” said Dr. Fred Chang, SVP and GM from AOI. “With the rapid build-out of greenfield AI networks and the expansion of existing infrastructure, there is an urgent need to reduce network power dissipation and cost of ownership, while ensuring overall network robustness. AOI’s new 400G and 800G designs, featuring the Credo Seagull 452 and Dove 800 DSPs, take advantage of the latest technologies available on the market to address these challenges.”


Zayo launches Waves on Demand across Europe with Ciena

Zayo is launching a new Waves on Demand service leveraging its 400G fully enabled lit-fibre network in Europe. This allows customers to turn up services across Europe within just five working days, significantly reducing service delivery intervals from the industry standard of 30+ days.

Zayo’s Waves on Demand, powered by Ciena’s WaveLogic 5 technology, enables customers to efficiently scale bandwidth on dedicated, pre-provisioned circuits, reaching up to 800G line side capacity. With Zayo’s Wavelength offering, customers can access industry-leading speed and pricing, including the best available in high capacity services, with service activation in just five working days. This rapid implementation model has been deployed on key connectivity corridors between Frankfurt, London, Amsterdam, Paris, Marseille, Bordeaux, and Dusseldorf. Coupled with Zayo’s metro network, this capability can be extended to key access locations across Western Europe. Zayo expects to expand this capability to Dublin, Manchester, and Milan by the end of this year.

Michael Katz, VP for Product & Technology at Zayo Europe, said: “At Zayo, we’re focused on addressing our customers’ needs now and into the future by connecting what’s next. We know that turning up Wave capacity can be a long and tedious process, leaving businesses with subpar services that impact their ability to deliver for their customers. Our new Waves on Demand offering will provide our customers with seamless service that is efficient, scalable, and cost-effective, meaning they can react in a matter of days to their business and customers’ needs.”

Will Rhodes, Director EMEA Carrier Managed Service Partners at Ciena, said: “The industry is crying out for faster turn-up of services. Getting high-capacity connectivity in several days will help to improve Zayo’s wholesale and enterprise customers’ agility and efficiency. This step-change will align better with end-user needs and digital experience.”


Dell'Oro: Explosive growth for AI servers

Strong demand for AI servers drove the revenues for GPU and custom accelerators by 224 percent in 2023, according to a new report from Dell'Oro Group. In 2024, in addition to continued growth momentum for AI infrastructure, we expect a broad recovery for the server and storage system component market as some of the hyperscale cloud service providers enter an expansion cycle, and enterprise IT spending gradually unfreezes.

"The accelerator market is set to maintain strong growth momentum as the hyperscale cloud service providers deploy next-generation AI infrastructure for larger training models and increasing inferencing demands," stated Baron Fung, Senior Research Director at Dell'Oro Group. "Enterprise adoption of AI applications is rising, driving the need for infrastructure in both public cloud and private data centers. With major product updates every two years, this market demands frequent infrastructure replacements, leading to higher capital investments. Furthermore, the recovery in the general-purpose computing and storage component market is underway after an extended correction. Fourth-generation CPUs from Intel and AMD, expected to ship in high volumes this year, will enable end-users to realize benefits in server efficiency and footprint optimizations. Pricing for commodities such as memory and storage drives will also rise this year, as supplies normalize, and from the deployment of new server architectures," explained Fung.

Additional highlights from the 4Q 2023 Data Center IT Semiconductors and Components Quarterly Report:

  • NVIDIA led in server and storage system component revenues in 2023, driven by GPU accelerators, followed by Intel and Samsung. Accelerator revenues surpassed CPU revenues for the first time in 2023, reflecting a shift towards accelerated computing.
  • GPU revenue is forecasted to achieve strong double-digit growth in 2024. Although NVIDIA currently dominates this market, potential challenges arise from new competitive offerings by AMD and Intel, coupled with the emergence of custom accelerators by major cloud service providers.
  • Smart NIC revenues grew more than 50 percent in 2023, driven by strong hyperscale adoption for both AI and non-AI use cases.

Coherent, Juniper and Marvell team on 800ZR

Juniper Networks is collaborating with Coherent Corp. and Marvellfo demo 800ZR.

The set-up features the Juniper Networks PTX10002-36QDD Packet Transport Router, Coherent Corporation’s 800ZR transceiver and the 800G Marvell Orion coherent digital signal processor (DSP).

  • Marvell Orion is an 800G coherent digital signal processor (DSP) for small form factor pluggable modules, delivering up to 800 Gbps of bandwidth for optical connections up to 1,000km and 400Gbps of bandwidth at up to 2,000km. The solution reduces cost per bit and power per bit by 30 percent compared to previous generation pluggable modules, while providing operators a path to substantially reduced capital and operational expenses.
  • Coherent Corporation’s 800G Digital Coherent Optics (DCO) transceiver utilizes its own integrated coherent transmitter and receiver Optical Sub-Assembly (IC-TROSA), enabling transmit optical output power of 0dBm in the dense QSFP-DD form-factor. The high optical output power enables IP-over-DWDM in DCI and metro/regional networks with transmission distances up to 1000 km at 800G.
  • The Juniper Networks 800G PTX10002-36QDD Packet Transport Router represents a next-generation, cloud-optimized routing platform, built upon Juniper's Express 5 ASIC. It offers unmatched operational simplicity, power efficiencies and enhanced performance, supporting dense 100GbE, 400GbE and 800GbE for scalability across various WAN and data center use cases, particularly in anticipation of significant capacity expansion driven by the rapid deployment of AI/ML clusters. It boasts 28.8 Tbps capacity.


Alphawave Semi and InnoLight collaborate on LPO + PCIe 6.0

Alphawave Semi and InnoLight Technology are demonstrating low power, low latency linear pluggable optics (LPO).

Operating at 64 Gbps per lane, this demonstration showcases Alphawave Semi’s PCIe 6.0 subsystem (Controller + PHY) with InnoLight’s LPO OSFP optics.

The Alphawave Semi PCIe Controller + PHY subsystem is an extremely power-efficient, low-latency, and highly reliable interface IP, having been built off PAM4 SerDes IP. It is available on the most advanced technology nodes. In AI and high-performance computing (HPC) systems where memory performance is key, the Alphawave Semi PCIe 6.0 Subsystem can be expanded to support CXL 3.0 with a PCIe and CXL Controller and enable memory coherency at the data center level.

Osa Mok, Chief Marketing Officer at InnoLight, said, “We are pleased with the collaboration between Alphawave Semi and InnoLight, which demonstrates PCIe 6.0 optical connectivity that offers both low power consumption and low latency, tailored to meet the demanding requirements of high-bandwidth and high-density AI networks. Alphawave Semi’s PAM4 DSP physical layer technology, leveraged with InnoLight’s LPO OSFP, empowers hyperscalers to implement next-generation disaggregated, composable networks.“

Tony Chan Carusone, CTO at Alphawave Semi, said, “We see our collaboration with InnoLight as a significant driver for accelerating the adoption of PCIe technology over optics. The combination of InnoLight’s market-leading optical solution with Alphawave Semi’s robust DSP technology showcases how PCIe technology over optics can enable our customers to massively scale connectivity for their compute and data processing AI workloads.”


Nokia tunes its WaveSuite optical network automation platform

Nokia has enhanced WaveSuite optical network automation platform by providing insight into and control of customer networks with a growing roster of applications targeting specific use cases to help them optimize their network and operations, scale network capacity and monetize network assets.

Key enhancements:

  • New support within the Service Enablement application for network operators to better integrate BSS (business support systems) billing with optical network functions. Network operators will be able to fulfill optical service requests to wholesale partners and their end subscribers, and provide them with individualized, real-time KPI assurance metrics. It enables them to offer differentiated services, such as latency-aware layer one services and spectrum-as-a-service, creating new revenue opportunities in a formerly untapped market. 
  • Machine learning into the Health and Analytics application to use network intelligence to support optical fiber sensing, which will help network operators detect disturbances to the fiber optic cable within their networks that may produce outages. Unlike other approaches that require additional hardware, the innovative WaveSuite solution takes advantage of polarization data provided by the coherent digital signal processor at the heart of Nokia Photonic Service Engines, such as PSE-6s. It allows network operators to proactively identify, prepare for and respond to events in the network that may affect network performance before they occur, decreasing down time and promoting better adherence to service level agreements with customers. 
  • Combining network planning with network provisioning, significantly reducing the effort required to scale network capacity by 33%. This enables automated flow-through provisioning by synchronizing between existing physical networks and planned network designs, eliminates errors in the transition between planning and provisioning, and significantly reduces the time and human resources network operators need to dedicate to scaling their optical networks. This enhancement was analyzed in a Jan 2024 Analysys Mason report where the research firm quantified the benefits of optical network automation. 

Ravi Parmasad, Optical Network Automation Leader at Nokia, said: "We continue to work with leading service providers to help them transform their optical networks with automation. Nokia's WaveSuite platform enables our customers to leverage data-driven insights and intelligent actions to optimize their network performance, reduce operational costs and accelerate service delivery. With WaveSuite, we are bringing the benefits of automation to the optical domain and empowering our customers to build more agile and resilient networks for the future.”


Cisco appoints Martin Lund as EVP, Common Hardware Group

Martin Lund has joined Cisco as EVP, Common Hardware Group, reporting to Chuck Robbins.

Previously, Lund was Corporate Vice President of Microsoft’s Azure for Operators, where he was responsible for delivering Azure-based solutions for public and private 5G, packet core, voice, and AI operations. Before being acquired by Microsoft, Martin was Chief Executive Officer of Metaswitch Networks, a cloud native communications software company. In addition to his 12 years at Broadcom, where he was instrumental in building the Network Switching brand and growing the business to more than $1 billion, Martin has also held leadership positions at companies such as Cadence Design Systems and Intel.

In addition, Eyal Dagan, who has previously led Cisco's Common Hardware Group since 2020, will now serve as EVP, Strategic Projects, also reporting to Chuck Robbins.

Cisco noted that Eyal and Martin have a friendship and partnership going back 20 years. In 2009, as the SVP and GM of Broadcom’s Network Switching business, Martin helped acquire Dune Networks, a company co-founded by Eyal.

Monday, March 25, 2024

Linear Pluggable Optics Consortium gets underway

A new Linear Pluggable Optics Multi-Source Agreement (LPO MSA) is underway to develop the specifications for networking equipment and optical modules.

The founding participants of the LPO MSA encompass a notable array of companies such as Accelink, AMD, Arista, Broadcom, Cisco, Eoptolink, Hisense, Innolight, Intel, MACOM, NVIDIA, and Semtech Corporation.

The initial target of the MSA is an optimized optical interconnect with LPO modules on both ends of the link. The LPO MSA specifications will define the electrical and optical requirements to ensure interoperability between multiple vendors of networking equipment and optics modules.

“There is an urgent need to reduce the network power consumption for AI and other high-performance applications,” said Mark Nowell, LPO MSA Chair. “LPO materially reduces power consumption both for the module and the system while maintaining a pluggable interface, providing the economics and flexibility that customers need for high-volume deployments.”

“The LPO MSA will enable a multi-source ecosystem for LPO optics modules which is essential for wide industry adoption,” said Andreas Bechtolsheim, Co-Chair LPO MSA. “We look forward to supporting the LPO MSA specifications.”

Marvell demos 3D Silicon Photonics (SiPho) Engine

Marvell is demonstrating a 3D Silicon Photonics Engine designed for next-generation AI clusters and cloud data centers. The new device packs 32 channels of 200G electrical and optical interfaces for connecting next-generation AI clusters and cloud data centers at multi-terabit speeds. 

The 200 Gbps device delivers 2x the bandwidth, 2x the input / output (I/O) bandwidth density, and 30% lower power per bit compared to comparable devices with 100 Gbps electrical and optical interfaces.

“Advances in optical interconnect technology are necessary to realize the promise of AI and accelerated infrastructure,” said Dr. Loi Nguyen, executive vice president and general manager of Cloud Optics at Marvell. “Our 3D SiPho engine is designed to enable higher bandwidth, higher I/O density, and lower power optical interconnects that scale cloud service providers’ infrastructure to meet the growing bandwidth needs of emerging AI services and applications.” 

TeraSignal intros CMOS Re-Driver for 800G LPO

TeraSignal has begun sampling a novel intelligent 400G (4x100G) PAM4 modulator driver with digital link training and link monitoring for 800G linear pluggable optical (LPO) modules. 

TeraSignal’s new TS8401/02 intelligent modulator driver addresses the limitations of traditional LPO modules, which unlike DSP-based re-timers, have lacked the capabilities for automatic adaptation, digital link monitoring and link training, making them challenging to integrate, interoperate, and deploy efficiently. By leveraging the existing microcontroller (MCU) in the LPO module, the TS8401/02 automates essential adjustments and introduces link diagnostics and link accountability, transforming LPO modules into adaptive link optimizing systems.

TeraSignal says its re-timer facilitates a closed-loop system that not only identifies but also rectifies link issues in real time, ensuring uninterrupted, high-quality data transmission across the board. The advanced CMOS implementation reduces power consumption by as much as 50% compared to existing Silicon Germanium (SiGe) solutions and significantly boosts link reliability and performance through statistical link monitoring and adaptive control capabilities.

“TeraSignal’s intelligent CMOS solution for LPO modules redefines high-speed interconnects, merging low latency and power efficiency with unmatched link performance measurement capabilities, diagnostics, and quick time to market,” said Dr. Armond Hairapetian, founder and CEO of TeraSignal. “We are committed to delivering IRD solutions with superior performance over both fiber and copper interconnects. By incorporating intelligence within the re-driver, we have introduced a new class of AI-centric CMOS devices that offer DSP-like features with significantly lower power and the lowest possible latency, setting a new industry standard for high-speed interconnects.”

Product Highlights

  • Intelligent 400G (4x100G) PAM4 modulator driver with integrated digital eye monitoring and link training.
  • Fully programmable linear output driver and equalizer for unparalleled signal integrity.
  • Automatic gain control supporting a wide range of input and output voltages.
  • More than 50% lower power consumption compared to non-CMOS drivers, significantly reducing operating costs and environmental impact.
  • Compact form factor and integration-ready design for seamless deployment in QSFP-DD/OSFP optical modules.

Additional TS8401/02 Benefits

  • Lower Power Consumption: Leveraging advanced design techniques in CMOS, the TS8401/02 operates at significantly lower power levels than existing DSP and SiGe-based solutions, marking a new standard in energy efficiency.
  • Reduced Latency: Due to their discrete-time nature, DSP-based re-timers must slow down the data traffic to perform the required data conversion, serialization/deserialization, and filtering functions. By eliminating the latency-causing building blocks in the continuous-time signal path, TS8401/02 significantly reduces data transmission latency.
  • Digital Eye Monitor: The on-board digital sampling scope in the TS8401/02 provides unparalleled flexibility and real-time insights into link performance, ensuring optimal signal integrity and reliability.
  • Link Training with Host SerDes: The TS8401/02 supports seamless integration and automatic configuration with host systems, further streamlining the deployment and operation of high-speed optical links.

    The TS8401 features wire bond pads, while the TS8402 offers the same capabilities in a flip-chip version.

New Photonics debuts Transmitter-on-Chip PIC

NewPhotonics introduced its second generation photonic integrated circuit with integrated optical equalizer for high-throughput optical interconnects. The enhanced transmitter-optimized chip offers breakthrough minimal latency and power performance at 800 GBps and 1.6 TBps for linear receive optics (LRO) and linear drive pluggable optics (LPO) applications in the data center.

The NewPhotonics NPG102 PIC is an octal and quad parallel single mode (PSM) transmitter for 8x and 4x 200G PAM4 transceivers. The silicon photonics chips monolithically integrate elements of the transmitter including laser, modulator and an optical equalizer enabling consistent optical link performance and extended distance. The best-in-class transmitter design simplifies system integration and improves OEM manufacturing yield and transceiver reliability.

The optical equalizer located closest to TP2 facilitates signal dispersion compensation that enables consistent optical link performance and interoperability. Optical signal processing enables >1000x lower latency resulting in lower system pJ/bit.

Key Highlights:

  • Performance consistency, improved distance, lower latency and power
  • Improved transceiver yields and extended reliability
  • 3.9W low power consumption
  • Simplifies system assembly with flip-chip package

"AI continues to drive data center processing needs of large language models (LLMs) ramping requirements for low power and low latency fiber connectivity," said Doron Tal, SVP and GM Optical Connectivity at NewPhotonics. "The differentiating utility of linear drive and optical equalization technology enables LPO modules to operate in all ports with different insertion losses ensuring interoperability between different host vendors and enabling a low power, low latency DSP-free alternative."

Intel integrates 224Gbps SerDes with NewPhotonics' Engine

Intel and NewPhotonics announced a milestone achievement: the integration of Intel new 224Gbps electrical SerDes design with NewPhotonics' Photonics Engine, resulting in an end-to-end direct modulation electrical-to-optical link utilizing PAM4 modulation.The electrical-to-optical link incorporates a wide bandwidth photodetector (PD) and Modulator, supporting the end-to-end 224Gbps data transfer."We are thrilled to have achieved the successful integration...

Corning intros Multifiber Pushlok connectors

Corning introduced its Multifiber Pushlok Technology for simplifying and accelerating fiber deployments.

Multifiber Pushlok is a “stick-and-click” connector technology thatallows operators to deploy more fiber in tighter spaces – an essential consideration for increasing data usage. It is a key feature in a new line of Evolv solutions, which take complicated splicing tasks out of the field to help installers connect homes and businesses more efficiently.

The technology enhances three new Corning Evolv solutions:

  • Evolve Assemblies with Multifiber Pushlok Technology add flexibility to network architecture with a variety of sizes and density options. Multifiber Pushlok’s tactile and audible feedback simplifies the cable assemblies’ installation.
  • Evolve Terminals with Multifiber Pushlok Technology include a new “stubless” version that reduces packaging material by up to 30% per assembly, allowing for up to 45% more product per shipping pallet.  
  • Evolv FlexNAP with Multifiber Pushlok Technology is “preconnectorized” to fit operators’ customer-specific locations, reducing their reliance on skilled labor and delivering cost savings of at least $25 per home, compared to traditional splice methods. The new system on RPX cable fits into 1.25-inch ducts, surpassing legacy solutions constrained to 2-inch ducts. This results in up to a 50% reduction in carbon footprint through minimized duct material usage.* The system also features a new built-in, locatable dust cap, which allows operators to swiftly locate the system for buried deployments and simplify network-troubleshooting efforts.

“We’re working alongside our customers to address new and ongoing challenges, such as cost constraints, labor shortages, and aggressive deployment timelines,” said Bob Whitman, vice president of market development, Carrier Networks, Corning Optical Communications. “Our Multifiber Pushlok Technology will help network operators bring the benefits of high-speed connectivity to more lives around the world – faster and more easily.”