Wednesday, June 9, 2021

Google's Firmina cable to link North and South America

Google announced plans to build a 12-fiber pair subsea cable linking North and South America via the Atlantic.

The Firmina cable system, which will be designed, manufactured and installed by SubCom, will run from the East coast of the United States to Las Toninas, Argentina, with additional landings in Praia Grande, Brazil and Punta del Este, Uruguay. 

The Firmina system will use SubCom's 18kV power technology, which will enable Firmina to be the world's longest cable capable of maintaining operations with single-end feed power, in the event there is a far-end fault. 


Manufacture of the cable and equipment will take place at SubCom’s recently expanded manufacturing campus in Newington, NH, USA during 2021 and early 2022, with main lay installation operations scheduled for summer 2022. The system is expected to be ready for service by the end of 2023.

In a blog post announcing the cable, Bikash Koley, VP Global Networking, Google Cloud stated: “As people and businesses have come to depend on digital services for many aspects of their lives, Firmina will improve access to Google services for users in South America. With 12 fiber pairs, the cable will carry traffic quickly and securely between North and South America, giving users fast, low-latency access to Google products such as Search, Gmail and YouTube, as well as Google Cloud services.”

The cable is named after Maria Firmina dos Reis (1825 - 1917), a Brazilian abolitionist, considered Brazil’s first novelist, whose 1859 novel, Úrsula, depicted life for Afro-Brazilians under slavery.

Including Firmina, Google now has investments in 16 subsea cables, such as Dunant, Equiano and Grace Hopper, and consortium cables like Echo, JGA, INDIGO, and Havfrue.

https://cloud.google.com/blog/products/infrastructure/announcing-the-firmina-subsea-cable

Avicena develops highly parallel optical interconnect for chips

Avicena, a start-up based in Mountain View, California, unveiled a highly parallel optical interconnect technology targeting up to 10 meters reach for chip-to-chip interconnects in distributed computing, processor-to-memory disaggregation, and other advanced computing applications. 

Avicena's LightBundle is based on arrays of novel GaN high-speed micro-emitters, leveraging the microLED display manufacturing ecosystem, and is fully compatible with high performance silicon ICs. The company has demonstrated an array of 200 CROME devices with a pitch of 30μm coupled to an array of PDs with a multi-core imaging fiber. Avicena says individual lanes show excellent performance characteristics up to a data rate of 10Gbps over the full temperature range of -40°C to 150°C. This extrapolates to an aggregate link bandwidth of 2Tbps for 200 lanes with a bandwidth density of 10Tbps/mm2.

“All of this is changing with the recent advances in optical emitter technology driven by advances in the display industry,” says Bardia Pezeshki, founder and CEO of Avicena. “We have developed very high-performance optical transmitters based on emitter technology from the display industry. These innovative devices would have been impractical just a few years ago. Our optimized devices and materials support 10Gbps links per lane over -40°C to +150°C temperature with excellent reliability. We refer to our new optical sources as Cavity-Reinforced Optical Micro-Emitters or CROMEs. We connect CROME arrays with CMOS compatible PDs using multi-core fiber bundles to create massively parallel interconnects with 1000s of parallel lanes over a reach of up to 10m. We call this new class of optical interconnect the Avicena LightBundleTM.”

The company says the parallel nature of its LightBundle technology is well-matched to parallel chiplet interfaces like AIB, HBI, and BoW, and can also be used to extend the reach of standard compute interconnects like PCIe, NVLink, and multi-channel G/DDR memory links with low power and low latency.

Corning showcases fiber innovations

Corning is showcasing its new fiber and cable innovations at this week's virtual OFC. Highlights:

  • MiniXtend cable line. This 192-fiber micro cable with binderless FastAccess® technology will leverage Corning’s new SMF-28 Contour fiber with 190-micron outer diameter. The compact MiniXtend XD cable provides installers with greater flexibility in tighter spces without sacrificing fiber density or deployment speed – enabling 33% more fibers per duct compared to legacy MiniXtend HD cable and 70% faster cable access compared to traditional cable designs.
  • RocketRibbon cable portfolio, including the world’s first armored cables with Corning’s FastAccess technology – offering ease of installation coupled with innovative ribbon protection. The innovation is offered in 288-fiber, 432-fiber, and 864-fiber variations. The compact designs enable up to twice as much fiber per duct and 60% faster cable access than existing solutions, while maintaining the benefits of a backward-compatible and proven industry-standard ribbon design.
  • SMF-28 ULL fiber, Corning’s newest long-haul terrestrial fiber, features advanced bend capability and designed for the most challenging long-haul and high-data-rate networks. Compared to standard single-mode fibers, it enables up to twice the capacity and reach at data rates of 800G and greater, at a lower cost per bit.
  • The latest innovations in Corning’s data center portfolio, including Corning CleanAdvantage Technology and MTP PRO Connecter with Push-Pull boot, which provide high-performance data centers with faster, cost-effective deployments and a streamlined path to revenue growth.
  • Evolv Hardened Connectivity Solutions with Pushlok Technology, designed to simplify fiber deployment for all types of communications networks, enabling 5G-ready connectivity in the most congested of environments.
  • Everon Network Solutions, the complete enterprise-grade network solution built on optical technology, delivering reduced costs at installation and over the lifetime of a building. The Everon portfolio includes solutions for the fiber backbone, cellular connectivity, and 5G-ready capabilities all the way to the edge of the network, giving building operators the flexibility and scale to design for the future.
  • Corning’s newest solutions for original equipment manufacturers, showcasing expertise in bringing high-density optical connectivity directly to the chip.

“Reliable, high-speed connectivity has never been more vital – from access networks, to cloud data centers, to smart buildings, to the promise of 5G,” said Dr. Bernhard Deutsch, vice president and general manager, Corning Optical Fiber and Cable. “Optical fiber, cable, and connectivity are at the center of all these technologies. As bandwidth demands accelerate, Corning remains at the forefront of innovation for every segment of the network.”


Corning’s experts will offer insights on industry trends during workshops throughout the conference. Topics including co-packaged optics; the next 10 years of optical networks; power efficiency in submarine fiber systems; and the past, present, and future of fiber innovation.  

AWS launches Proton app delivery service for containers

Amazon Web Services announced the general availability of AWS Proton, an application delivery service that makes it easier for customers to provision, deploy, and monitor microservices for container and serverless applications. 

With AWS Proton, a customer’s infrastructure team creates standard application stacks defining the architecture, infrastructure resources, CI/CD (continuous integration and continuous delivery) pipeline, and observability tools—and then makes these stacks available to their developers. Developers can use AWS Proton’s self-service interface to select an application stack for use with their code. AWS Proton automatically provisions the resources for the selected application stack, deploys the code, and sets up monitoring so developers can begin building serverless and container applications without having to learn, configure, or maintain the underlying resources. There are no upfront commitments or fees to use AWS Proton, and customers pay only for the AWS services used to create, scale, and run their applications. 


“Customers have told us that while they love the operational benefits that container and serverless applications provide, it is incredibly challenging to scale these architectures across their organizations because of the many manual tasks involved in deploying apps that use microservices,” said Deepak Singh, VP, Compute Services, AWS. “AWS Proton brings together customers’ infrastructure as code, CI/CD pipeline, and observability into a single interface, so developers can quickly go from code in a repo to a production application. Developers rely on AWS Proton’s self-service capabilities to deploy code quickly and securely without having to become experts on each of the underlying services involved, while the central infrastructure team can be assured that the apps deployed by their developers using AWS Proton meet the standards they have set for their business.”

AWS Proton is now available in US East (N. Virginia), US East (Ohio), US West (Oregon), Europe (Ireland), and Asia Pacific (Tokyo) with additional region availability coming soon.

https://aws.amazon.com/proton


Xilinx debuts Versal AI Edge series processors

Xilinx introduced the Versal AI Edge series processors, boasting 4X the AI performance-per-watt versus GPUs and 10X greater compute density versus previous-generation adaptive SoCs.

Xilinx is positioning the new Versal AI Edge adaptive compute acceleration platforms (ACAPs) for a range of applications including: automated driving with the highest levels of functional safety, collaborative robotics, predictive factory and healthcare systems, and multi-mission payloads for the aerospace and defense markets. The portfolio features AI Engine-ML to deliver 4X machine learning compute compared to the previous AI Engine architecture and integrates new accelerator RAM with an enhanced memory hierarchy for evolving AI algorithms. These architectural innovations deliver up to 4X AI performance-per-watt versus GPUs and lower latency resulting in far more capable devices at the edge.

"Edge computing applications require an architecture that can evolve to address new requirements and scenarios with a blend of flexible compute processing within tight thermal and latency constraints,” said Sumit Shah, senior director, Product Management and Marketing at Xilinx. “The Versal AI Edge series delivers these key attributes for a wide range of applications requiring greater intelligence, making it a critical addition to the Versal portfolio with devices that scale from intelligent edge sensors to CPU accelerators.”

The Versal AI Edge series takes the production-proven 7nm Versal architecture and miniaturizes it for AI compute at low latency, all with power efficiency as low as six watts and safety and security measures required in edge applications. As a heterogeneous platform with diverse processors, the Versal AI Edge series matches the engine to the algorithm, with Scalar Engines for embedded compute, Adaptable Engines for sensor fusion and hardware adaptability, and Intelligent Engines for AI inference that scales up to 479 (INT4) TOPS2—unmatched by ASSPs and GPUs targeting edge applications—and for advanced signal processing workloads for vision, radar, LiDAR, and software defined radio.

Sampling is available to early access customers, with shipments expected during the first half of 2022.

https://www.xilinx.com/versal-ai-edge

Samsung announces 8nm RF chip architecture for 5G

Samsung Electronics introduced its newest radio frequency (RF) technology based on 8-nanometer (nm) process.

Samsung’s 8nm RF process technology is the latest addition to an already broad portfolio of RF-related solutions, including 28nm- and 14nm-based RF. The company notes shiptment of more than 500 million mobile RF chips for premium smartphones since 2017.

Samsung has developed a unique architecture exclusive to 8nm RF named RFextremeFET (RFeFET) that can significantly improve RF characteristics while using less power. In comparison to 14nm RF, Samsung’s RFeFET supplements the digital PPA scaling and restores the analog/RF scaling at the same time, thereby enabling high-performance 5G platforms. Compared to 14nm RF, Samsung’s 8nm RF process technology provides up to a 35-percent increase in power efficiency with a 35-percent decrease in the RF chip area as a result of the RFeFET architectural innovation.

“Through excellence in innovation and process manufacturing, we’ve reinforced our next-generation wireless communication offerings,” said Hyung Jin Lee, Master of Foundry Technology Development Team at Samsung Electronics. “As 5G mmWave expands, Samsung’s 8nm RF will be a great solution for customers looking for long battery life and excellent signal quality on compact mobile devices.”

Marcatel deploys 600G Long-haul in Mexico with Ciena

Marcatel, a telecommunications carrier based in Mexico, is deploying Ciena’s coherent optical technology to support a new network connecting Querétaro, Mexico, to McAllen, Texas. The network, which serves content delivery providers and wholesale carriers, allows Marcatel to offer low-latency, high-capacity data center interconnect (DCI) services at speeds up to 600Gbps across long distances.


“Marcatel has long been a pioneer in the networking industry in Mexico, and we’re again raising the bar with the first network in the country to carry 600G waves for long-haul distances,” said Gustavo M. de la Garza Flores, CEO of Grupo Marcatel.

Marcatel installed Ciena’s flexible 6500 photonic line system and Waveserver 5 powered by WaveLogic 5 Extreme for multi-terabit capacity across a long-haul coherent optical network of more than 1000km, providing high fiber capacity without regeneration. With this deployment, Marcatel can instantly react to unpredictable traffic demands by scaling connectivity and capacity to support a wide array of applications and services—all while taking up less space in equipment rooms with a compact footprint.

http://www.ciena.com

Amazon activates its 3rd Availability Zone in the Beijing with Sinnet

Amazon Web Services is adding a third Availability Zone (AZ) to the China (Beijing) Region operated by Sinnet.


Fow AWS, an AZ encompasses one or more discrete data centers in separate facilities, each with redundant power, networking, and connectivity. With this launch, both AWS Regions in China offer three AZs.


MACOM and MaxLinear pair PAM4 DSPs and 100G/lane TIAs

MACOM and MaxLinear announced a collaboration to assure interoperability of MaxLinear’s PAM4 DSPs and MACOM’s 100G/lane transimpedance amplifiers (TIAs) for 100G/lambda applications. 

The collaboration includes pairing the MxL93516, MaxLinear’s second-generation 100G Snowmass DSP, with MACOM’s low power MATA-05817 100G TIA to achieve exceptional sensitivity and a low BER floor. This configuration is ideal for 100G-DR1/FR1/LR1 QSFP28 applications for Data Center and front haul applications.

“Our customers are requesting proven high-performance solutions,” said Marek Tlalka, Senior Director, High-Performance Analog, at MACOM. “This collaboration offers the industry a compelling performance proof point and a potential path to shorter design cycle times for 100G, 400G and 800G module manufacturers.”

“The market for 100G/lambda transceivers continues to grow and performance and time to market are critical for our customers,” said Drew Guckenberger, Vice President of MaxLinear’s High-Speed Interconnect Group. “The pairing of our second generation power optimized Snowmass 100G DSP with integrated driver with MACOM’s high-performance TIAs is an ideal solution for customers, providing tight integration, proven high-performance and low-power capabilities.”

The companies plan to continue collaborating on 400Gbps and 800Gbps applications utilizing the recently announced MaxLinear 5nm Keystone DSPs and MACOM’s extensive portfolio of four channel 400Gbps TIAs.

https://www.macom.com/products/product-detail/MATA-05817

Go!Foton offers "Any Density” optical patch panel

 Go!Foton announced the general availability of NEMO (Network Enhanced Manageable Optics), its lightweight patch panel. Powered by PEACOC, the company’s flagship fiber management platform which introduced spreadable adapters to fiber connectivity management, NEMO targets traditional bulkhead-style fiber panels and enables substantially higher port counts in any fixed space when compared with legacy installations.

“NEMO’s spreadable adapters allow superior access and secure fiber management for any number or type of connection. It is ideal for today’s high-capacity and data-rate intensive telecommunication and data communication networks,” said Go!Foton CTO Dr. David Z. Chen.

“Uniquely in the industry, NEMO allows untooled error-free access to all fiber connections for even the largest hands, and from both sides of the rack, while minimizing insertion loss,” Dr. Chen continued. “With its support for an extensive variety of connector types—including LC, SC, SN, CS, MDC, and MPO, and its ability to accommodate any cable size from 1.2mm-3.5mm, NEMO is the ideal connectivity solution for optical network operators confronting the challenge of growing densification.”

https://www.gofoton.com/hot-new-gfa-items/

Riverbed appoints Dan Smoot as President and CEO

Riverbed appointed Dan Smoot as President and CEO.

Smoot, who has more than 30 years of experience holding top leadership roles including at Salesforce, Cisco and VMware, joined Riverbed in June 2018. At Riverbed, he has held several executive leadership roles, including most recently COO and leading the company’s Worldwide Sales, Channels and Alliances, Sales Operations and Customer Experience teams.

Smoot was previously Executive Vice President of Global Partner Sales at Salesforce, managing the Worldwide Partner Sales organization. He was also Executive Vice President of Market Readiness at Salesforce, where he created a global organization to deliver critical go-to-market operations. Prior to Salesforce, Smoot worked at VMware as Senior Vice President of Global Customer Operations, overseeing advanced technology sales, global services, renewals, global channels and alliances, OEM/ISV sales and go-to-mark strategy. Before VMware, he spent more than 12 years in a variety of senior leadership roles at Cisco.

“Dan has been a stand-out leader at Riverbed and brings tremendous industry experience having held senior leadership roles at some of world’s leading technology companies,” said Robert ‘Tre’ Sayle, Riverbed Board Member and Thoma Bravo Partner (Riverbed is a portfolio company of Thoma Bravo). “Dan has an unwavering commitment to Riverbed’s customers and partners, an exceptional track record driving operational excellence and business results, and is a strong leader of people and culture. Dan is the right leader to take Riverbed to the next level and drive profitable growth for our business, and continued innovation to support customers and partners. I also want to thank our previous CEO, Rich McBee, who helped lead Riverbed during a critical time, including setting a strong foundation for the future and navigating the Company through a global pandemic.”

http://www.riverbed.com

Tuesday, June 8, 2021

Open XR Forum MSA gets underway

A new Open XR Forum has been established to advance the development of XR optics-enabled products and services, accelerate the adoption of coherent point-to-multipoint network architectures, and drive standardization of networking interfaces to ensure ease of multi-vendor interoperability and an open, multi-source solution ecosystem.

The Open XR Forum will function as a multi-source agreement (MSA) working group for XR optics, the industry’s first point-to-multipoint coherent pluggable transceiver technology. XR optics provides a solution that can dramatically simplify network architectures, reduce operating costs, and enable a dynamic and on-demand capacity allocation while paving the way for a network architecture that is compatible across multiple generations of optical technologies. Network operator studies have demonstrated that XR optics can provide a reduction in the total cost of ownership by as much as 70 percent.

Initial members include Verizon, Lumen Technologies, Windstream, Liberty Global and Infinera. 

http://www.openxrforum.org

Infinera introduces XR Optics-powered coherent pluggables

Infinera introduced a suite of coherent optical pluggables based on its XR optics technology and designed to seamlessly address point-to-point and point-to-multipoint transport applications from the network edge to the core. Infinera’s new suite of vertically integrated ICE-XR pluggables will support a range of transport rates, including 100G, 400G, and 800G, and utilize industry-standard form factors such as QSFP-28, QSFP-DD, and CFP2 to enable...

Australia's NBN Co tests Infinera's XR optics

NBN Co, Australia’s wholesale open-access broadband provider, completed a successful proof of concept of Infinera's XR optics-based point-to-multipoint coherent optical technology. The trial demonstrated XR optics’ compatibility within the nbn Transit Network, a DWDM network spanning over 65,000 km across Australia.Infinera said the trial highlighted the ability of XR optics to dramatically simplify transport network architecture while reducing CapEx...


ADVA's CoreChannel transmits 800G over 1,600-km

ADVA introduced its CoreChannel technology with the power to transmit 800 Gbps over more than 1,600 km using 140GBd sub-carrier technology, representing an improvement in reach of over 50% compared with 95GBd technologies.

ADVA says the new CoreChannel sled for FSP 3000 TeraFlex platform is now being trialed by major carriers and R&E network operators. Compatible with deployed infrastructure, the new solution delivers industry-leading SNR requirements making it resilient to transmission impairments. Along with the bandwidth flexibility and fractional QAM capabilities of the TeraFlex terminal, this ensures 800 Gbps transport with highest efficiency, highest performance and lowest cost. Featuring multi-rate client support, the CoreChannel™ sled can transport any mix of 400GbE, 100GbE and 10GbE services, providing customers with flexibility and protecting their investment as they migrate to higher speeds.

“Network operators across the planet have been harnessing our FSP 3000 TeraFlex to unlock the full potential of their fiber assets. Our compact, plug-and-play terminal offers a simple and cost-efficient way to maximize the capacity of optical infrastructure. Today we’re making it even more powerful with the introduction of our CoreChannel™ sled. Customers can now transport data at 400Gbit/s and 800Gbit/s over longer distances than ever before with a solution that delivers the lowest possible cost per bit per kilometer,” said Christoph Glingener, CTO, ADVA. “Our FSP 3000 TeraFlex™ CoreChannel™ technology offers an enormous boost to operators transmitting services across countries and continents. What’s more, long-haul transport with TeraFlex™ CoreChannel™ is even more efficient and reliable thanks to its superior link signal-to-noise (SNR) requirements and higher network performance.”

https://www.adva.com/en/newsroom/press-releases/20210608-adva-launches-teraflex-corechannel-for-industrys-longest-800g-reach


Ayar Labs demos terabit link for Co-Packaged Optics and chip-to-chip

Ayar Labs demonstrated the industry’s first Terabit per second Wavelength Division Multiplexing (WDM) optical link with its TeraPHY optical I/O chiplet and SuperNova multi-wavelength optical source. 

The demonstration shows a fully functional TeraPHY chiplet with 8 optical ports running error free without Forward Error Correction (FEC) for a total bandwidth of 1.024 Tbps and at less than 5 pJ/bit energy efficiency. 

“This is yet another industry first demonstration of our technology which promises to transform computing and extend the benefits of Moore’s Law,” said Charles Wuischpard, CEO, Ayar Labs. “Digital transformation is being driven by cloud, connectivity, artificial intelligence, and intelligent edge; our roadmap of products and custom solutions is tailored to serve these high-volume market segments. We remain on track to deliver on several customer commitments by end of year.”

Ayar Labs also demonstrated the industry’s first multi-wavelength, multi-port optical source with 64 addressable wavelengths. Named SuperNova, it offers eight times the number of wavelengths compared to commercially available pluggable products. The SuperNova is also the first product compliant with the optical source specifications of the CW-WDM MSA, an industry consortium that drives standards for advanced optical communication and computing applications that require a leap in performance, efficiency, cost, and bandwidth scaling. The multi-wavelength source leverages laser technology from MACOM, a leading supplier of lasers and optoelectronics for telecommunications and data center applications.

“MACOM’s novel laser arrays are ideal for silicon photonics and co-packaged optics,” said Stephen G. Daly, President and CEO of MACOM. “As a promoter-level member of the CW-WDM MSA, we are excited to collaborate with Ayar Labs on the next generation of optical devices based on the MSA’s new specification.”

The details of this industry-first demonstration will be presented as a postdeadline paper by Dr. Mark Wade, Co-Founder and CTO of Ayar Labs, on Friday, June 11 at 9:15 a.m. PDT at the Optical Fiber Communication Conference (OFC) 2021. OFC postdeadline papers represent the latest and most advanced technical achievements in the field.

(available via OFC conference web site for registrants)

https://ayarlabs.com

Ayar Labs hits key milestone for chip-to-chip optical connectivity

 Ayar Labs demonstrated its patented monolithic electronic/photonic solution on Globalfoundries' next-gen photonics solution based on its 45nm platform. The companies said this industry-first demonstration is a key milestone in providing chip-to-chip optical connectivity at scale. The two companies began working together in 2015 with a commitment to collaborate and commercialize differentiated silicon photonics solutions for greenfield...

Lockheed Martin Ventures invests in Ayar Labs

Lockheed Martin Ventures has made a strategic investment in Ayar Labs, a start-up that is developing  monolithic in-package optical I/O (MIPO) solution for applications that require high bandwidth, low latency and power efficient short reach interconnects. Financial terms were not disclosed. Ayar Labs publicly demonstrated its monolithic electronic photonic TeraPHY chiplet at the Supercomputing 2019 conference and is now working with select...

Ayar Labs has been selected as Intel’s optical I/O solution partner for their recently awarded DARPA PIPES (Photonics in Package for Extreme Scalability) project. The PIPES project aims to develop integrated optical I/O solutions co-packaged with next generation FPGA/CPU/GPU and accelerators in Multi-Chip Packages (MCP) to provide extreme data rates (input/output) at ultra-low power over much longer distances than supported by current technology....



CW-WDM MSA releases Multi-Wavelength Optical Laser Sources spec

The CW-WDM MSA (Continuous-Wave Wavelength Division Multiplexing Multi-Source Agreement) Group released its first official specification for 8, 16, and 32 wavelength optical sources.

The Multi-Source Agreement (MSA) defines laser sources covering three spans in the O-band wavelength grids. These lasers can be used as optical sources for multiple applications such as high- density co-packaged optics, machine learning, and optical computing. Each wavelength grid is defined with enough technical detail to ensure interoperability across the range of applications. The MSA also outlines measurement methods required for standards compliance.

Key Highlights of the MSA Specifications:

  • Frequency assignments for 8, 16 and 32 wavelength grid configurations with multiple grid spacings
  • Two physical configurations, including a modular optical source with each output port carrying a single wavelength, and an integrated optical source with each output port carrying all the wavelengths
  • A range of output power classes targeting applications that require very low and very high output levels
  • Definitions of optical parameters and measurement methods, including relative intensity noise, side mode suppression ratio (SMSR), and linewidth
  • Fixed and flexible wavelength grid configurations to support deployment in a variety of environmental conditions

“The CW-WDM MSA just published an 8, 16, and 32 wavelength O-band laser source standard for use in optical transceivers for emerging advanced datacom applications,” commented Chris Cole, Chair of the MSA. “Laser sources are the critical component in optical communications and having an agreed set of grids creates great opportunities for transceiver and laser suppliers to develop innovative products.”

https://cw-wdm.org/cw-wdm-msa-consortium-releases-new-specification-for-multi-wavelength-optical-laser-sources/

Marvell intros Atlas 50Gbps PAM4 DSP chipset

Marvell announced its Atlas 50Gbps PAM4 DSP chipset which integrates transimpedance amplifiers (TIAs) and laser drivers in the CMOS package.

The integration of the new Atlas chipset, which is based on Marvell’s Polaris 50G PAM4 DSP family,  reduces supply chain complexity and power consumption by up to 25%. Atlas is also the industry’s first PAM4 DSP available to wire bond directly with lasers and photodiodes, further simplifying the assembly process for optical module integrators, effectively reducing both product lead times and time to market. Sampling is underway.

Atlas PAM4 DSP Key Features:

  • Leverages Polaris-class quad-channel DSP core, providing world-class performance and signal integrity while driving faster time to market and product yields.
  • Uses existing Polaris DSP software, minimizing software development expense for module developers.
  • Comprehensive suite of performance monitoring tools to reduce bring-up time and support mission-mode telemetry critical for hyperscale deployments.
  • Integrated laser drivers support both single- and multi-mode interconnect applications.
  • Suitable to support a wide range of module form factors including OSFP, QSFP56, QSFP-DD and AOCs.
  • Broad range of IEEE standard support including 100G-xR2, 200G-xR4 and 400G-xR8.
  • Manufactured in mainstream CMOS process technology optimized for high volume production to deliver a high-performance, low-power PAM4 DSP solution.

“Disaggregated data centers define today’s new cloud computing paradigm, and this trend is driving hyper-connectivity within the data center, putting PAM4 data center interconnects at the heart of the network,” said Xi Wang, VP of Marketing, Optical Connectivity at Marvell. “The new Atlas PAM4 DSP targets this burgeoning, high-volume data center market as well as cost-sensitive applications at the network edge.”

“Marvell understands the requirements of today’s hyperscale data center operators and has delivered a highly integrated, high-performance PAM4 DSP.

https://www.marvell.com/ 

Infinera’s ICE6 selected for PEACE submarine cable system

Infinera's ICE6 optical engine solution will be deployed on the Mediterranean Segment of the Pakistan & East Africa Connecting Europe (PEACE) submarine cable system. 

Infinera said it was selected following extensive technology evaluation and application analysis of its ICE6 solution, including its combination of ultra-high baud rates and an advanced modulation technique known as long-codeword probabilistic constellation shaping. Both capabilities are using the latest 7-nm ASIC technology, highly optimized algorithms and Infinera’s second-generation Nyquist subcarriers.

The high-speed, 15,000-km PEACE submarine cable system offers high-capacity, low-latency routes connecting China, Europe and Africa. In addition to France, the cable will land in Malta, Cyprus, Egypt, Djibouti, Kenya, Pakistan and other countries and regions, with onward terrestrial connectivity to China.

Mr. Sun Xiao Hua, Chief Operating Officer, PEACE Cable International Network Co., Limited, said, “We are committed to being a leading international submarine cable operator, providing customers with an open, flexible and carrier-neutral service that enables them to choose any device that they prefer. This collaboration with Infinera is just the beginning. Welcome aboard.”

Mr. Haitham Zahran, Vice President – EMEA Subsea Cable Systems, PCCW Global, said, “PEACE submarine cable system is specifically optimized for maximum capacity per fiber pair and our advanced network design enables us to continue to embrace new technologies in order to meet growing bandwidth demand over these important routes. We are hopeful that it will soon be the highest-performing open cable system connecting Asia, Europe and Africa.”

Mr. Nick Walden, Senior Vice President of Worldwide Sales, Infinera, said, “Infinera has always excelled in providing solutions that help network operators extract the greatest amount of value from their submarine assets. ICE6 enables PCCW Global and PEACE Cable to offer industry-leading, open, scalable and flexible services to their customers over the PEACE submarine cable system, as demand across these regions continues to grow dramatically.”

http://www.peacecable.com


Fastly's global CDN suffers configuration error

 Fastly, which operates a global content delivery network, experienced service configuration issue that triggered disruptions across its POPs worldwide. Service disruptions continued for about three hours, impacting traffic from top websites such as CNN, NYTimes, Amazon, Reddit, and others.

According to the company's website, the incident affected: Asia/Pacific (Auckland (AKL), Brisbane (BNE), Dubai (FJR), Hong Kong (HKG), Melbourne (MEL), Osaka (ITM), Perth (PER), Singapore (SIN), Sydney (SYD), Tokyo (HND), Tokyo (TYO), Wellington (WLG), Singapore (QPG), Tokyo (NRT)), South America (Buenos Aires (EZE), Bogota (BOG), Curitiba (CWB), Rio de Janeiro (GIG), Santiago (SCL), Sāo Paulo (CGH), Sāo Paulo (GRU), Lima (LIM)), North America (Ashburn (BWI), Ashburn (DCA), Ashburn (IAD), Ashburn (WDC), Atlanta (FTY), Atlanta (PDK), Boston (BOS), Chicago (CHI), Chicago (MDW), Chicago (ORD), Chicago (PWK), Columbus (CMH), Columbus (LCK), Dallas (DAL), Dallas (DFW), Denver (DEN), Houston (IAH), Jacksonville (JAX), Kansas City (MCI), Los Angeles (BUR), Los Angeles (LAX), Los Angeles (LGB), Miami (MIA), Minneapolis (MSP), Minneapolis (STP), Montreal (YUL), New York (LGA), Newark (EWR), Palo Alto (PAO), Phoenix (PHX), Portland (PDX), San Jose (SJC), Seattle (SEA), St. Louis (STL), Toronto (YYZ), Vancouver (YVR)), South Africa (Cape Town (CPT), Johannesburg (JNB)), India (Chennai (MAA), Mumbai (BOM), New Delhi (DEL)), and Europe (Amsterdam (AMS), Copenhagen (CPH), Dublin (DUB), Frankfurt (FRA), Frankfurt (HHN), Helsinki (HEL), London (LCY), London (LHR), London (LON), Madrid (MAD), Manchester (MAN), Marseille (MRS), Milan (MXP), Oslo (OSL), Paris (CDG), Stockholm (BMA), Vienna (VIE), Munich (MUC)).

https://status.fastly.com/incidents/vpk0ssybt3bj

MACOM intros two-chip analog solution for 200G and 400G

MACOM announced the availability of its two-chip analog solution for 200G and 400G short reach modules (QSFP, OSFP, QSFP-DD) and Active Optical Cable (AOC) applications in the Data Center. The chipset delivers better than IEEE standard specified bit error rate (BER) performance and is compliant with Open Eye MSA eye mask specifications.

The chipset consists of two integrated combo devices:

  • MATA-38044, a 4x56Gbps PAM-4 CDR and TIA
  • MALD-38045, a 4x56Gbps PAM-4 CDR and VCSEL driver

The MATA-38044 is a four channel 28 GBaud linear PAM-4/NRZ TIA with automatic gain control and integrated CDRs. The MALD-38045 is a four channel 28 GBaud PAM-4/NRZ VCSEL driver with an integrated CDR designed to directly modulate VCSELs in the transmit path of optical modules. Each of the four channels can operate independently. Both devices consume industry leading low power and can operate at Ethernet, Fibre Channel and InfiniBand PAM-4 and NRZ data rates up to 56Gbps.


This two-chip analog solution provides an alternative to traditional DSP-based technology. It has lower power consumption, lower latency, lower cost and requires a smaller footprint when compared to a DSP-based solution.

https://www.macom.com/

ADVA supplies FSP 3000 open line system for TIP's proof-of-concept

ADVA supplied its FSP 3000 open line system (OLS) and Ensemble Controller network management and SDN domain controller system for a Telecom Infra Project (TIP) open optical networking proof of concept (PoC) at this week's virtual OFC.

The PoC showed how multi-vendor network components can be managed by a unified software-defined networking (SDN) control system. Specifically designed for easy interoperability, ADVA’s FSP 3000 OLS and Ensemble Controller ensured seamless integration into the open environment, including third-party terminals. 

Conducted by TIP’s Converged Architectures for Network Disaggregation & Integration (CANDI) sub-group, the PoC proved that all components of an open optical network can be operated by a single SDN controller. By demonstrating unified management of the OLS, open terminals and optical planning tool, the demo paves the way for true open networking and a new model of infrastructure operations with operators taking over network integration and optimization processes. 

“This PoC is an important milestone in the journey to fully open and disaggregated optical networking. It offers new levels of visibility and a way to manage the entire multi-vendor environment. With the power to seamlessly integrate components from different suppliers, network operators can confidently deploy best-in-class infrastructure that meets their exact requirements,” said Christoph Glingener, CTO, ADVA. “Our FSP 3000 OLS was central to the demo. It’s a truly open solution that gives operators freedom to build multi-supplier architectures and then evolve and optimize each network layer separately. With our Ensemble Controller also playing a major role – providing network management and programmatic SDN control – this demo highlights the value of technology like ours that supports interoperability and disaggregation.”

https://www.adva.com/en/newsroom/press-releases/20210607-tip-harnesses-adva-for-successful-trial-of-end-to-end-open-optical-network