Thursday, April 3, 2025

Keysight Unveils AI Data Center Scaling Architecture

Related stories

Centrilogic & CrewAI Partner to Boost AI Adoption in Canada

Centrilogic, a global provider of IT transformation solutions, has...

Adaptive Security Raises $43M to Fight AI Cyber Threats

Adaptive Security, the leading provider of AI-powered social engineering...

Denso’s Machinery & Tools Adopts CADDi for AI-Powered Ops

CADDi, an AI data platform driving the digital transformation...

GrowthLoop Launches AI-Powered Compound Marketing Engine

Agentic technology transforms marketing into a growth engine, shortening...

Opsera Secures $20M to Advance AI DevOps & Efficiency

Opsera reports 200% revenue growth since raising Series A+...
spot_imgspot_img

Enables AI infrastructure providers to emulate and optimize all aspects of the data center, from the physical layer through the application layer

Keysight Technologies, Inc. has introduced the Keysight Artificial Intelligence (KAI) architecture, an innovative portfolio of end-to-end solutions designed to help data centers scale AI processing capabilities. By enabling real-world AI workload emulation, KAI validates AI cluster components, ensuring seamless system-level interoperability, performance optimization, and efficiency insights. This comprehensive approach allows operators to enhance system performance and detect potential issues that traditional component-level testing might overlook.

Scaling AI infrastructure requires thorough validation throughout the design and deployment phases. Every component—chips, cables, interconnects, switches, servers, and GPUs—must be rigorously tested at both individual and system levels. While physical layer testing remains essential, full-stack workload emulation provides deeper insights, allowing organizations to unlock maximum AI performance faster. This accelerates AI capacity expansion and optimizes return on investment in large-scale AI clusters.

Key Benefits of the KAI Architecture

The KAI architecture is designed to assist AI providers, semiconductor manufacturers, and network equipment suppliers in three critical areas:

  • Accelerating Design: Debug high-speed digital designs and ensure compliance with PCIe, DDR, and CXL industry standards.
  • Accelerating Development: Validate component-level performance, including high-speed interconnects, cables, and chipsets, while ensuring optimal system-level workload performance.
  • Accelerating Deployment & Operations: Use end-to-end emulation to optimize data center performance, minimizing the risk of workload failures before production deployment.

Also Read: EDGNEX Data Centers partners with Hyperco

Comprehensive AI Data Center Solutions

The KAI architecture includes the newly launched KAI Data Center Builder alongside four specialized solution suites, covering every stage of AI data center development—from pre-silicon simulation to post-deployment testing and troubleshooting.

  • KAI Data Center Builder: Simulates large-scale AI workloads with measurable fidelity, enhancing system performance, predicting component failures, and streamlining operations.
  • KAI Compute: Enhances high-speed digital designs and advances AI chip development with AI-ready tools such as electronic design automation software, bit-error ratio testers, oscilloscopes, and arbitrary waveform generators.
  • KAI Interconnect: Ensures scalable, high-speed optical and electrical connectivity up to 1.6T using AI-ready tools, including sampling oscilloscopes, photonic power meters, and interconnect testers.
  • KAI Network: Benchmarks AI network performance, detects bottlenecks, and optimizes workload distribution with AI-ready tools such as network traffic emulators and distributed traffic generators.
  • KAI Power: Maximizes power efficiency across data center components with AI-ready tools, including oscilloscopes, power rail probes, and electronic design automation solutions.

Industry Expert Insights

Alan Weckel, Founder and Technology Analyst at 650 Group, stated: “Accelerating design and deployment of next-generation AI/ML ASICs is key to unlocking the market as customers move from foundational training to agentic models. AI interconnect through scale-up, scale-out, and frontend networks will drive record 800GE and 1.6T port shipments over the next several years with one of the fastest innovation cycles to ever occur in the industry. The Keysight Artificial Intelligence architecture and KAI Data Center Builder solution will help customers scale and adapt to this new market opportunity.”

Ram Periakaruppan, Vice President and General Manager of Network Test & Security Solutions at Keysight, added: “Scaling AI data centers requires more than component-level validation. Interoperability, performance, and efficiency are system-wide metrics that can only be measured under real-world network conditions. Keysight’s AI solutions integrate our deep experience in traffic emulation, component, and network compliance validation, and the latest industry standards to emulate every aspect of data center performance: compute, network, interconnect, and power to ensure AI infrastructure meets evolving demands.”

Subscribe

- Never miss a story with notifications


    Latest stories

    spot_img