Here's the page we think you wanted. See search results instead:

Discover three new products that will revolutionize your AI data center

A new era for AI data centers

AI workloads push massive data across millions of components. Small issues cascade, resulting in up to 50% idle processors and reduced AI performance.

Watch the video to learn how KAI Data Center Builder emulates every processor, interconnect, and network to maximize AI data center performance.

Test every data path in your system at scale

KAI Data Center Builder emulates real-world AI workloads to test every data path in your system at scale. Visualize AI data flows through every processor, every interconnect, every network element and pinpoint performance bottlenecks.

Explore the use case to learn more.

How to Emulate AI Data Center Workloads

How to Emulate AI Data Center Workloads

Emulating AI data center workloads requires a consistent, repeatable and scalable testing methodology. Learn how to optimize infrastructure for AI training performance.

Gain deeper product insights from our solution briefs, white papers, and application notes.

Solution Briefs 2025.04.02

Keysight AI Data Center Builder

Keysight AI Data Center Builder

This solution brief introduces how Keysight AI (KAI) Data Center Builder addresses these challenges by accelerating AI / ML networks and input/output innovation cycles. This innovative software uses high-density traffic load appliances that support 100-800GE ports to emulate AI hosts and realistic AI / ML traffic patterns. You will learn how KAI Data Center Builder can help speed up releases, lower costs, simplify setup and maintenance, and improve overall performance.

2025.04.02

White Papers 2025.03.10

How AI / ML Networks Differ from Traditional Networks

How AI / ML Networks Differ from Traditional Networks

Artificial intelligence (AI) and machine learning (ML) are driving big changes in data center networks — including new requirements, traffic patterns, and technologies. Discover how Keysight solutions help optimize AI networks.

2025.03.10

Application Notes 2025.03.10

AI Fabric Test Methodology

AI Fabric Test Methodology

In the context of Artificial Intelligence (AI) systems, particularly Large Language Models (LLMs), a network fabric refers to the underlying infrastructure that connects and enables communication between various components or nodes within the system. The network fabric is crucial in facilitating efficient and effective AI system operation, including LLM training.

2025.03.10

Reports 2024.10.30

Tolly Test Report: Dell PowerEdge and Broadcom AI Performance Benchmarking

Tolly Test Report: Dell PowerEdge and Broadcom AI Performance Benchmarking

Artificial Intelligence (AI) places new and unprecedented demands on servers and the network fabric. During the AI training process, massive amounts of data need to be exchanged among servers. Dell commissioned Tolly to benchmark the IPv6 RDMA offload performance of the Broadcom 57508 Dual-Port 100 Gb/s Ethernet Controller (LAN adapter) running in the Dell PowerEdge R760xa Rack Servers by testing with Keysight AI benchmarks.

2024.10.30

Test your 1.6T interconnects and networks

Keysight 1.6T Network Performance Tester emulates Ethernet traffic at speeds up to 1.6 Tbps. Test reliability, stability, and interoperability of chips, transceivers,  interconnects, and networking equipment.

Explore the use case to learn more.

How to Validate Ethernet Interconnects in Data Centers

How to Validate Ethernet Interconnects in Data Centers

Validating Ethernet interconnects in data centers requires precise testing for reliability and performance. Learn how to ensure high-quality data transmission and error correction.

Gain deeper product insights from our solution briefs, white papers, and application notes.

Application Notes 2025.03.27

Validating 800GE and 1.6TE Interconnects

Validating 800GE and 1.6TE Interconnects

Resilient 800GE and 1.6T Ethernet data center interconnects play a vital role in maximizing AI network performance. While they promise greater data rates and lower latencies, these transceivers will introduce higher modulation formats, more sensitive components, and increased susceptibility to impairments. To meet demanding performance requirements and ensure future scalability, interconnect vendors must validate both compliance with evolving standards and seamless interoperability with existing data center infrastructure. This application note explores the significant opportunities emerging in the rapidly growing AI/ML data center market and addresses the key challenges vendors face, such as maintaining high signal integrity, optimizing power consumption, and ensuring compatibility with modern network architectures. It also outlines a comprehensive set of test strategies, including rigorous performance testing, interoperability assessments, and adherence to industry standards. These strategies are crucial for the successful deployment of interconnects in AI data centers.

2025.03.27

Solution Briefs 2025.03.17

Interconnect and Network Performance Tester 1600GE

Interconnect and Network Performance Tester 1600GE

The Interconnect and Network Performance Tester 1600GE (INPT-1600GE) is a versatile high speed Ethernet platform, available in a portable, lightweight, office quiet benchtop or rackmount chassis. The platform validates performance and reliability from 200GE to 1600GE in a range of Ethernet devices and interconnects that use 212Gb/s electrical lane interfaces.

2025.03.17

White Papers 2025.04.14

Future-Proofing Networks with 1.6T Ethernet

Future-Proofing Networks with 1.6T Ethernet

This white paper delves into the key challenges of validating interconnect performance in the context of 1.6T Ethernet. It explores the growing demands on data centers, the limitations of existing network architectures, and the emerging technologies and testing methodologies required to ensure the reliable deployment of these high-speed networks.

2025.04.14

White Papers 2025.04.09

AI at Full Throttle

AI at Full Throttle

This white paper discusses the challenges and solutions related to high-speed Ethernet networks in the context of rapidly growing AI applications. It highlights the critical need for advanced networking technologies to support the increasing demands of AI workloads.

2025.04.09

Test your 1.6T optical transceivers

Keysight 1.6T Optical Transceiver Tester accelerates 1.6T transceiver development and manufacturing tests. Analyze signal integrity and validate industry compliance to maximize performance in any operating condition.

Explore the use case to learn more.

How to Optimize 1.6T Optical Transceiver Manufacturing Tests

How to Optimize 1.6T Optical Transceiver Manufacturing Tests

Scaling 1.6T optical transceiver production requires fast, efficient transmitter dispersion and eye closure quaternary (TDECQ) measurements. Learn to accelerate TDECQ measurements with test optimization software, optical switching, and 224 Gb/s oscilloscopes to maximize yield and efficiency.

Gain deeper product insights from our solution briefs, white papers, and application notes.

White Papers 2025.03.11

800G / 1.6T Data Center Transceiver Test

800G / 1.6T Data Center Transceiver Test

This white paper delves into AI data center trends, addressing how hyperscale data center architects and operators must scale networks from 400G to 800G / 1.6 terabit (1.6T) Ethernet to meet these demands. This guide explores key strategies to increase interconnect bandwidth, ensure interoperability and reliability, and reduce test time, power, and cost.

2025.03.11

Application Notes 2025.03.25

TDECQ Measurement Update

TDECQ Measurement Update

This paper discusses how the measurement of TDECQ has evolved and improved, and what users of Keysight systems should expect for accuracy and consistency for a wide variety of test scenarios.

2025.03.25

White Papers 2023.06.08

TDECQ Part 1: Making Accurate and Repeatable Compliance Measurements

TDECQ Part 1: Making Accurate and Repeatable Compliance Measurements

TDECQ is the acronym used to describe “transmitter dispersion and eye closure quaternary”. This is a power penalty metric that describes how much extra power is required from a transmitter, relative to an ideal transmitter, to compensate for both non-ideal transmitter waveforms and the impact of chromatic dispersion. It is similar to the transmitter dispersion penalty (TDP) metric used in NRZ optical communications standards. It was developed in the IEEE 802.3bs standard and improved in the IEEE 802.3cd standard. It is a convenient way to specify the performance of a PAM4 transmitter, as it aggregates many transmitter attributes into a single metric, and the metric directly indicates the impact a transmitter’s imperfections will have on overall system-level performance.

2023.06.08

White Papers 2023.11.27

TDECQ Part II Manufacturing Test Recommendations

TDECQ Part II Manufacturing Test Recommendations

This document gives an overview of the adoption of TDECQ in manufacturing tests, and how to efficiently automate FlexDCA for making TDECQ measurements.

2023.11.27

See How Juniper Networks Boosted AI Network Performance

“With Keysight, we are able to replicate the real data center environment — specifically for AI data center requirements, such as RoCEv2.”

Mahesh Subramaniam, Director of Product Management, JUNIPER NETWORKS

Expand your AI data center expertise

AI data center, AI data centers, AI datacenter, KAI innovations, Keysight AI Data Center Builder

Watch Keysight's AI data center showcase on demand

See product overviews of all 3 of our latest AI data center test solutions, driving state of the art in data speeds, throughput, and compute performance.  

Watch webinar on demand
1.6T, transceiver, AI data center solutions

Take an Interactive Tour of Keysight's Optical Transceiver Test

Ready to optimize optical transceiver test efficiency with our optical test optimization software?

See for yourself how they can simplify configurations of optimal TDECQ, OMA, and Outer Extinction Ratio measurements.

Take a product tour
1.6T, transceiver, AI data center solutions

Course: Introduction to AI Data Center Emulation and Optimization

In this expert-led course, you'll explore how to emulate realistic AI workloads, stress-test data center performance under real-world conditions, and quickly pinpoint system bottlenecks across layers.

Discover why traditional component-level testing falls short to meet AI demands and how next-gen emulation solutions can help you build resilient, high-throughput AI data center networks.

Start learning today
AI data center white paper, AI data center cluster, white paper on AI datacenter

Improving scalability in AI data center clusters

Can your network infrastructure scale to handle the complex, high-traffic AI training workloads?

Read this white paper to learn AI data center cluster scaling, critical network challenges, and strategies to ensure scalable and reliable networks for your organization’s AI ambitions.

Read white paper
benchmark AI data center, AI data center performance , benchmark collective AI datacenter

Benchmarking Collective Operations

Measuring or benchmarking the network performance in an AI data center cluster can help organizations identify opportunities to optimize and improve overall throughput without additional hardware costs. This white paper explains the operation of AI collectives, defines terminology, and reviews the most common metrics associated with benchmarking AI networks.

Read white paper

Ready to scale your AI data center with Keysight?

Talk to us today