Tuesday, September 12, 2023

Nokia debuts its next-gen IP access, aggregation and edge routers

Nokia unveiled its 7730 Service Interconnect Router (SXR) family of next-generation IP access, aggregation and edge platforms based on its new FPcx  network processing unit (NPU), 112G SerDes, and SFP-DD optics.

The new, fully programmable FPcx processor, which is implemented in 7nm silicon, enables network security features such as in-line filtering for DDoS mitigation and line-rate MACsec/ANYsec encryption. The chip also features 7-level hierarchical QoS, queueing and a hierarchical memory design with superior scalability for services, tunnels, queues, counters. 

The FPcx supports 1GE to 400GE interfaces with flexible 4x10G, 4x25G, 2x100G, 4x100G and 1x400G break-out options.

Nokia says the NPU can accommodate new services and standards as they emerge.

The new 7730 Service Interconnect Router (SXR) family includes four fixed form-factor and two modular platform variants, with a mix of port options ranging from 1GE to 400GE.  Backplanes range from 3.6 Tbps to 5.6 Tbps of system capacity.

The platforms support stringent synchronization demands with highly accurate timing enabled by integrated dual-band Global Navigation Satellite System (GNSS) receivers. 7730 can support time-sensitive packet transport and fronthaul networks with a Class C-based Telecom Boundary Clock.

The rollout brings focus to the Nokia SR Linux network operating system (NOS), which embraces cloud-native design principles, and the Nokia Network Services Platform (NSP) automation suite.

Ken Kutzler, Vice President of IP Routing Hardware at Nokia, said: “As service providers and network builders scale their IP networks for increased capacity, they require the ability to confidently deliver deterministic, secure and reliable services with easy, efficient service evolution and network operations – all with solid TCO savings. Nokia’s new family of routers enables them to get the benefits of service routing available today at the IP edge/core into the IP access and aggregation networks. We do this without performance compromise while ensuring the latest requirements are met.”


Accton intros 51.2 Tbps switch with 64x800G ports

Taiwan-based Accton Technology introduced a series of 800G-optimized products that can provide Ethernet-based fabric and Virtual Output Queue (VoQ) based fabric for AI/ML workloads. 

The announcement includes a 2RU 51.2 Tbps Ethernet-based fabric system with 64x800G ports, a 4RU 51.2 Tbps Ethernet-based fabric system with 128x400G ports, and a VoQ-based fabric Distributed Disaggregated Chassis (DDC) architecture with a 128x800G Network Cloud Fabric (NCF) engine.

Accton said its 51.2 Tbps Broadcom StrataXGS Tomahawk 5 series-based system with 64x800G or 128x400G ports provides a high-radix, deployment-friendly Ethernet-fabric. The 2RU system design is packed in a compact form factor with power and fan tray redundancy to achieve five-nines high availability and a wide environmental operating range for data center cloud applications. 

The platform comes in two variants, OSFP800 or QSFP-DD800 interface options, for flexible deployment supporting passive copper DAC on all ports and long-distance ZR+ optics. The 4RU 51.2 Tbps Tomahawk 5 series-based 128x400G system provides high-radix 200G/400G QSFP112 (compatible with QSFP56 and QSFP28) connectivity to accelerators and compute nodes in a flat architecture that reduces latency and required power, which enables networks to be scaled-out sustainably.

Accton’s Broadcom StrataDNX Jericho3-AI and StrataDNX Ramon3 systems are designed to form a VoQ-based fabric DDC Network Cloud Cluster that is optimized to support AI/ML applications. The Broadcom Jericho3-AI based Network Cloud Packet Forwarder (NCP) supports a 14.4 Tbps full-duplex switching and features 18x800G OSFP network interface ports and 20x800G OSFP fabric interface ports. The Broadcom Ramon3-based Network Cloud Fabric (NCF) engine supports a 51.2 Tbps switching capacity. The cell-based switching eliminates the Ethernet overhead and can effectively load balance all fabric links to build an efficient and high-availability DDC cluster.

With Accton’s 6RU, dual-Ramon3, 128x800G (dual-51.2 Tbps) fabric port switch and 2RU, Jericho3-AI, 18x800G (14.4 Tbps leaf) Ethernet network interface port switch, customers can build a 32K-GPU 800G/400G AI/ML cluster with a two-stage DDC network architecture to enable 400G GPU clusters now and migrate to 800G GPU clusters later with a software upgrade without replacing the switches. This provides excellent cost saving and investment protection that reduces capital spending while providing the flexibility for customers to “pay as you grow” without large upfront cost to enable AI and ML applications.

Michael K.T. Lee, Sr. Vice President of Research and Development Center, at Accton said, “Broadcom’s Tomahawk 5 and Roman3/Jericho3-AI form the building blocks of Accton’s 800G-optimized products for next-generation AI/ML clusters. We are proud of Accton's engineers who have excelled in the innovative design and development of these groundbreaking 800G systems. The significant increase in radix and throughput offered by the 800G AI fabric together with the reduced power consumption delivers a performance that will push the boundaries of efficiency in AI clusters.”


Enfabrica raises $125M for its 8 Tbps Accelerated Compute Fabric

Enfabrica,  a start-up based in Mountain View, California, closed a $125 million Series B financing round for its converged networking and memory fabric silicon and software.

The company says the new capital will be deployed to advance the production of itsAccelerated Compute Fabric Switch (ACF-S) devices and solutions, which complement GPUs, CPUs and accelerators to solve critical networking, I/O and memory scaling problems in data center AI and high-performance computing clusters. 

Earlier this year, Enfabrica emerged from stealth mode to announce its new class of AI infrastructure interconnect chips called Accelerated Compute Fabric devices. Enfabrica’s ACF devices deliver unmatched scalability, performance and total cost of ownership for distributed AI, extended reality, high-performance computing and in-memory database infrastructure.

Enfabrica’s 8 Terabit/second ACF switching system, which directly bridges and internetworks GPUs, CPUs, and memories at scale with native, multi-port 800-Gigabit-Ethernet networking, is now accessible to customers for pre-order.

Enfabrica’s first, production-grade ACF switching system powered by Enfabrica silicon, produced in collaboration with partners, is an 8 Tbps platform enabling direct-attach of any combination of GPUs, CPUs, CXL-attached DDR5 memory and SSD storage to high-performance, multi-port 800-Gigabit-Ethernet networks. ACF switching systems incorporate Enfabrica’s high-performance ACF-S silicon having 100 percent standards-compliant interfaces and host networking software stack components running on standard Linux kernel and userspace interfaces. The ACF-S chip offers multi-port 800 Gigabit Ethernet networking and high-radix PCIe Gen5 and CXL 2.0+ interfaces. 

The company envisions composable AI fabrics of compute, memory and network resources that scales from a single system to tens of thousands of nodes. The design provides uncontended access to >50X DRAM expansion over existing GPU networks via ComputeExpressLink (CXL) bridging.

“The fundamental challenge with today’s AI boom is the scaling of infrastructure,” said Rochan Sankar, Enfabrica CEO and co-founder. “There’s no denying the transformative value that AI delivers to a multitude of economic sectors. But there is a critical need to bridge the exploding demand to the overall cost, efficiency, and ease of scaling AI compute, across all customers seeking to take control of their distributed AI infrastructure and services. Much of the scaling problem lies in the I/O subsystems, memory movement and networking attached to GPU compute, where Enfabrica’s ACF solution shines.”

Infinera's ICE-X coherent pluggables comply with Buy America

Infinera confirmed that its ICE-X intelligent coherent pluggables and compound semiconductor components  will be manufactured in the U.S. and compliant with the Build America, Buy America requirements recently released by the Commerce Department for the Broadband Equity, Access, and Deployment (BEAD) program. 

Infinera operates an optical compound semiconductor fabrication facility in California and an advanced testing and packaging facility in Pennsylvania.

Infinera’s suite of vertically integrated ICE-X intelligent coherent pluggables, based on the optical compound semiconductor indium phosphide, provide network operators the performance, scale, and efficiency critical to drive down network operating costs and to enhance service agility. These solutions are a critical part of network infrastructure and are optimized for metro, middle-mile, and access and aggregation networks that enable operators to expand their broadband infrastructures while helping them address Build America, Buy America requirements for the BEAD program.

“Infinera is a global leader in the development and production of advanced optical semiconductors, operating one of the most advanced U.S.-based compound optical semiconductor fabrication facilities for monolithically integrated photonic integrated circuits. We remain committed to investing in and keeping optical semiconductor fabrication in the U.S.,” said David Heard, Infinera CEO. “By leveraging our vertically integrated U.S.-based development, fabrication, and manufacturing capabilities, we can do our part to enhance national security and improve supply chain resiliency for important semiconductor technologies.”


OpenZR+ MSA Group adds 400G 8QAM and higher transmit power

The OpenZR+ Multi-Source Agreement (MSA) Group published the OpenZR+ Rev 3.0 specification which defines a higher performing 400G 8QAM mode as well as a higher transmit power mode.

The new specification is designed to further expand the application space for a coherent solution in small form factor pluggable modules, such as form factors defined by the QSFP-DD MSA and OSFP MSA.

“The OpenZR+ MSA is once again paving the way towards multi-vendor interoperability of small form factor pluggables, which are key for enabling network operators to support their growing bandwidth demands in a more cost-efficient manner,” said Atul Srivastava, Co-chair of the OpenZR+ MSA Group.

“With this Rev 3.0 specification, the MSA is continuing to expand the application space for OpenZR+ with new modes designed to address longer reaches and a wider range of network architectures,” said Tom Williams, Co-chair of the OpenZR+ MSA Group.

In the future, the OpenZR+ MSA plans to continue to expand membership in the organization with the goal of further collaborating on improvements to the specification that are in the best interest of the industry. The OpenZR+ MSA expects to play a key role in defining higher data rate operating modes for pluggable coherent interfaces as the industry looks toward 800G and beyond. 


Intel to sell 10% stake in the IMS Nanofabrication business to TSMC

Intel agreed to sell an approximately 10% stake in the IMS Nanofabrication business to TSMC. 

TSMC’s investment values IMS at approximately $4.3 billion, consistent with the valuation of the recent stake sale to Bain Capital Special Situations. 

Intel will retain majority ownership of IMS, which will continue to operate as a standalone subsidiary under the leadership of CEO Dr. Elmar Platzgummer. The transaction is expected to close in the fourth quarter of 2023.

Intel said Bain Capital's and TSMC’s investments provide IMS with increased independence and reinforce confidence in the significant opportunity ahead of IMS. This added autonomy will help IMS accelerate its growth and drive the next phase of lithography technology innovation to enable the industry’s transition into new patterning systems, such as high-numerical-aperture (high-NA) EUV.

IMS specializes in multi-beam mask writing tools required to develop advanced extreme ultraviolet lithography (EUV), which is broadly adopted in leading-edge technology nodes that enable the most demanding computing applications, such as artificial intelligence (AI) and mobile. 

Matt Poirier, senior vice president of Corporate Development at Intel, said, “This investment demonstrates the deep industry collaboration IMS is pioneering to advance critical lithography technology for leading-edge nodes, which will benefit the entire semiconductor manufacturing ecosystem. With enhanced independence, IMS will be well positioned to address the significant growth opportunity for multi-beam mask writing tools over the next decade and beyond.”