NVIDIA Introduces Spectrum-XGS Ethernet to Connect Distributed Data Centers Into Giga-Scale AI Super-Factories
NVIDIA (NASDAQ:NVDA) has unveiled Spectrum-XGS Ethernet, a groundbreaking scale-across technology designed to unite distributed data centers into unified, giga-scale AI super-factories. This innovation adds to the NVIDIA Spectrum-X Ethernet platform, enabling data centers to scale beyond single facilities and overcome traditional Ethernet infrastructure limitations.
The technology nearly doubles the performance of NVIDIA's Collective Communications Library through advanced features like auto-adjusted distance congestion control and precision latency management. CoreWeave will be among the first to implement this technology, connecting their data centers into a unified supercomputer.
Spectrum-XGS Ethernet is now available as part of the NVIDIA Spectrum-X Ethernet platform, which delivers 1.6x greater bandwidth density than traditional Ethernet solutions.
NVIDIA (NASDAQ:NVDA) ha presentato Spectrum-XGS Ethernet, una tecnologia rivoluzionaria pensata per scalare orizzontalmente e collegare data center distribuiti in un'unica super-fabbrica AI su scala giga. Questa novità arricchisce la piattaforma NVIDIA Spectrum-X Ethernet, permettendo ai data center di andare oltre i limiti delle singole strutture e superare le restrizioni delle infrastrutture Ethernet tradizionali.
La tecnologia quasi raddoppia le prestazioni della Collective Communications Library di NVIDIA grazie a funzionalità avanzate come il controllo della congestione con adattamento automatico alla distanza e una gestione precisa della latenza. CoreWeave sarà tra i primi ad adottarla, collegando i propri data center in un supercomputer unificato.
Spectrum-XGS Ethernet è ora disponibile come parte della piattaforma NVIDIA Spectrum-X Ethernet, che offre una densità di banda 1,6x superiore rispetto alle soluzioni Ethernet tradizionali.
NVIDIA (NASDAQ:NVDA) ha lanzado Spectrum-XGS Ethernet, una tecnología innovadora diseñada para escalar horizontalmente y unir centros de datos distribuidos en súper fábricas de IA a escala giga. Esta incorporación amplía la plataforma NVIDIA Spectrum-X Ethernet, permitiendo que los centros de datos superen las limitaciones de instalaciones individuales y de la infraestructura Ethernet tradicional.
La tecnología casi duplica el rendimiento de la Collective Communications Library de NVIDIA mediante funciones avanzadas como el control de congestión ajustado automáticamente según la distancia y una gestión precisa de la latencia. CoreWeave será uno de los primeros en implementarla, conectando sus centros de datos en un superordenador unificado.
Spectrum-XGS Ethernet ya está disponible como parte de la plataforma NVIDIA Spectrum-X Ethernet, que ofrece una densidad de ancho de banda 1,6x mayor que las soluciones Ethernet tradicionales.
NVIDIA (NASDAQ:NVDA)가 Spectrum-XGS Ethernet을 공개했습니다. 이는 분산된 데이터센터들을 하나의 통합된 기가 스케일 AI 슈퍼팩토리로 연결하도록 설계된 획기적인 확장형 기술입니다. 이 혁신은 NVIDIA Spectrum-X Ethernet 플랫폼에 추가되어 데이터센터가 단일 시설의 한계를 넘어 기존 이더넷 인프라의 제약을 극복할 수 있게 합니다.
이 기술은 거리 기반 자동 혼잡 제어와 정밀한 지연시간 관리 같은 고급 기능을 통해 NVIDIA의 Collective Communications Library 성능을 거의 두 배로 끌어올립니다. CoreWeave는 이 기술을 가장 먼저 도입하는 기업 중 하나로, 자사 데이터센터들을 통합된 슈퍼컴퓨터로 연결할 예정입니다.
Spectrum-XGS Ethernet은 이제 NVIDIA Spectrum-X Ethernet 플랫폼의 일부로 제공되며, 기존 이더넷 솔루션보다 1.6배 더 높은 대역폭 밀도를 제공합니다.
NVIDIA (NASDAQ:NVDA) a dévoilé Spectrum-XGS Ethernet, une technologie révolutionnaire conçue pour s'étendre à grande échelle et unir des centres de données répartis en usines d'IA giga-échelle. Cette innovation s'ajoute à la plateforme NVIDIA Spectrum-X Ethernet, permettant aux centres de données de dépasser les limites des installations individuelles et d'éliminer les contraintes des infrastructures Ethernet traditionnelles.
La technologie double presque les performances de la Collective Communications Library de NVIDIA grâce à des fonctions avancées comme le contrôle de congestion ajusté automatiquement à la distance et une gestion précise de la latence. CoreWeave fera partie des premiers à la déployer, connectant ses centres de données en un supercalculateur unifié.
Spectrum-XGS Ethernet est désormais disponible dans le cadre de la plateforme NVIDIA Spectrum-X Ethernet, qui offre une densité de bande passante 1,6x supérieure à celle des solutions Ethernet traditionnelles.
NVIDIA (NASDAQ:NVDA) hat Spectrum-XGS Ethernet vorgestellt, eine bahnbrechende, übergreifend skalierbare Technologie, die verteilte Rechenzentren zu einheitlichen Giga-Scale-AI-Superfabriken verbindet. Diese Neuerung erweitert die NVIDIA Spectrum-X Ethernet-Plattform und ermöglicht Rechenzentren, über einzelne Standorte hinaus zu skalieren und die Einschränkungen traditioneller Ethernet-Infrastrukturen zu überwinden.
Die Technologie verdoppelt nahezu die Performance von NVIDIAs Collective Communications Library durch fortschrittliche Funktionen wie automatisch an die Distanz angepasste Staukontrolle und präzises Latenzmanagement. CoreWeave wird zu den ersten gehören, die die Technologie einsetzen und ihre Rechenzentren zu einem einheitlichen Supercomputer verbinden.
Spectrum-XGS Ethernet ist jetzt als Teil der NVIDIA Spectrum-X Ethernet-Plattform verfügbar, die eine 1,6-fach höhere Bandbreitendichte als herkömmliche Ethernet-Lösungen bietet.
- Introduction of revolutionary scale-across technology for connecting distributed data centers
- Nearly doubles the performance of NVIDIA's Collective Communications Library
- Delivers 1.6x greater bandwidth density than off-the-shelf Ethernet solutions
- Early adoption by major player CoreWeave demonstrates market validation
- Immediate availability as part of the Spectrum-X Ethernet platform
- None.
Insights
NVIDIA's new Spectrum-XGS Ethernet enables massive multi-data center AI operations, strengthening their dominance in AI infrastructure.
NVIDIA has unveiled Spectrum-XGS Ethernet, a groundbreaking networking technology designed to connect geographically distributed data centers into unified AI super-factories. This innovation addresses a critical limitation in the AI infrastructure landscape - the physical constraints of single-facility data centers that currently cap computing power.
The technology introduces what NVIDIA calls "scale-across" capabilities, complementing traditional scale-up and scale-out approaches. While scale-up involves adding more powerful components to a single system and scale-out means adding more systems within a data center, scale-across connects entire data centers across cities or even continents.
What makes this technically significant is the solution to distance-related networking challenges. Standard Ethernet suffers from high latency, jitter, and unpredictable performance when spanning large distances. Spectrum-XGS includes adaptive algorithms that dynamically adjust to inter-facility distances, implementing advanced congestion control and precision latency management that nearly doubles the performance of NVIDIA's Collective Communications Library.
CoreWeave's early adoption is telling - they're positioning to create what amounts to a distributed supercomputer spanning multiple physical locations. This allows AI workloads to harness computing resources that exceed what any single facility could provide, potentially enabling more complex and larger AI models.
This release represents the third major networking announcement in NVIDIA's strategy, following Spectrum-X and Quantum-X silicon photonics switches. Together, these create a comprehensive networking fabric specifically optimized for AI workloads at unprecedented scale - further entrenching NVIDIA's position as the infrastructure provider of choice for advanced AI development.
CoreWeave to Deploy NVIDIA Spectrum-XGS Ethernet With Scale-Across Technology
PALO ALTO, Calif., Aug. 22, 2025 (GLOBE NEWSWIRE) -- Hot Chips -- NVIDIA today announced NVIDIA® Spectrum-XGS Ethernet, a scale-across technology for combining distributed data centers into unified, giga-scale AI super-factories.
As AI demand surges, individual data centers are reaching the limits of power and capacity within a single facility. To expand, data centers must scale beyond any one building, which is limited by off-the-shelf Ethernet networking infrastructure with high latency and jitter and unpredictable performance.
Spectrum-XGS Ethernet is a breakthrough addition to the NVIDIA Spectrum-X™ Ethernet platform that removes these boundaries by introducing scale-across infrastructure. It serves as a third pillar of AI computing beyond scale-up and scale-out, designed for extending the extreme performance and scale of Spectrum-X Ethernet to interconnect multiple, distributed data centers to form massive AI super-factories capable of giga-scale intelligence.
“The AI industrial revolution is here, and giant-scale AI factories are the essential infrastructure,” said Jensen Huang, founder and CEO of NVIDIA. “With NVIDIA Spectrum-XGS Ethernet, we add scale-across to scale-up and scale-out capabilities to link data centers across cities, nations and continents into vast, giga-scale AI super-factories.”
Spectrum-XGS Ethernet is fully integrated into the Spectrum-X platform, featuring algorithms that dynamically adapt the network to the distance between data center facilities.
With advanced, auto-adjusted distance congestion control, precision latency management and end-to-end telemetry, Spectrum-XGS Ethernet nearly doubles the performance of the NVIDIA Collective Communications Library, accelerating multi-GPU and multi-node communication to deliver predictable performance across geographically distributed AI clusters. As a result, multiple data centers can operate as a single AI super-factory, fully optimized for long-distance connectivity.
Hyperscale pioneers embracing the new infrastructure include CoreWeave, which will be among the first to connect its data centers with Spectrum-XGS Ethernet.
“CoreWeave’s mission is to deliver the most powerful AI infrastructure to innovators everywhere,” said Peter Salanki, cofounder and chief technology officer of CoreWeave. “With NVIDIA Spectrum-XGS, we can connect our data centers into a single, unified supercomputer, giving our customers access to giga-scale AI that will accelerate breakthroughs across every industry.”
The Spectrum-X Ethernet networking platform provides 1.6x greater bandwidth density than off-the-shelf Ethernet for multi-tenant, hyperscale AI factories — including the world’s largest AI supercomputer. It comprises NVIDIA Spectrum-X switches and NVIDIA ConnectX®-8 SuperNICs, delivering seamless scalability, ultralow latency and breakthrough performance for enterprises building the future of AI.
Today’s announcement follows a drumbeat of networking innovation announcements from NVIDIA, including NVIDIA Spectrum-X and NVIDIA Quantum-X silicon photonics networking switches, which enable AI factories to connect millions of GPUs across sites while reducing energy consumption and operational costs.
Availability
NVIDIA Spectrum-XGS Ethernet is available now as part of the NVIDIA Spectrum-X Ethernet platform.
Learn more about Spectrum-XGS Ethernet at Hot Chips.
About NVIDIA
NVIDIA (NASDAQ: NVDA) is the world leader in accelerated computing.
For further information, contact:
Alex Shapiro
Enterprise Networking
1-415-608-5044
ashapiro@nvidia.com
Certain statements in this press release including, but not limited to, statements as to: with NVIDIA Spectrum-XGS Ethernet, NVIDIA adding scale-across to scale-up and scale-out capabilities to link data centers across cities, nations and continents into vast, giga-scale AI super-factories; the benefits, impact, performance, and availability of NVIDIA’s products, services, and technologies; expectations with respect to NVIDIA’s third party arrangements, including with its collaborators and partners; expectations with respect to technology developments; and other statements that are not historical facts are forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, which are subject to the “safe harbor” created by those sections based on management’s beliefs and assumptions and on information currently available to management and are subject to risks and uncertainties that could cause results to be materially different than expectations. Important factors that could cause actual results to differ materially include: global economic and political conditions; NVIDIA’s reliance on third parties to manufacture, assemble, package and test NVIDIA’s products; the impact of technological development and competition; development of new products and technologies or enhancements to NVIDIA’s existing product and technologies; market acceptance of NVIDIA’s products or NVIDIA’s partners’ products; design, manufacturing or software defects; changes in consumer preferences or demands; changes in industry standards and interfaces; unexpected loss of performance of NVIDIA’s products or technologies when integrated into systems; and changes in applicable laws and regulations, as well as other factors detailed from time to time in the most recent reports NVIDIA files with the Securities and Exchange Commission, or SEC, including, but not limited to, its annual report on Form 10-K and quarterly reports on Form 10-Q. Copies of reports filed with the SEC are posted on the company’s website and are available from NVIDIA without charge. These forward-looking statements are not guarantees of future performance and speak only as of the date hereof, and, except as required by law, NVIDIA disclaims any obligation to update these forward-looking statements to reflect future events or circumstances.
© 2025 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, ConnectX and Spectrum-X are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with which they are associated. Features, pricing, availability and specifications are subject to change without notice.
A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/4b20de4d-8e6b-4c75-b19f-c8b030d0cf3f
FAQ
What is NVIDIA's new Spectrum-XGS Ethernet technology and what does it do?
How much performance improvement does NVIDIA's Spectrum-XGS Ethernet provide?
Who is the first major customer to adopt NVIDIA's Spectrum-XGS Ethernet?
When will NVIDIA's Spectrum-XGS Ethernet be available?
What are the key features of NVIDIA's Spectrum-XGS Ethernet?
