SANTA ROSA, Calif. - Keysight Technologies , Inc. (NYSE: NYSE:KEYS) and Advanced Micro Devices, Inc. (NASDAQ: NASDAQ:AMD) have announced a breakthrough in network performance benchmarking by introducing a novel testing methodology for mobile and 5G networks.
This development aims to address the growing demand for bandwidth and the need for a realistic evaluation of network infrastructure capable of supporting millions of high-bandwidth, low-latency devices.
The new benchmarking approach, which utilizes Keysight's software and AMD's 4th generation EPYC CPUs, is designed to assess the true scalability and power of CPUs under realistic conditions. The methodology focuses on generating real-world traffic to test the infrastructure's ability to handle the simultaneous demands of voice, video, and web sessions.
This innovative testing solution is particularly relevant for industries undergoing digital transformation and automation, such as smart manufacturing and smart energy grids. Traditional benchmarks often fail to measure the CPU's potential under the network's bandwidth constraints. Still, Keysight and AMD's integrated test case methodology seeks to provide a more accurate representation of a CPU's performance by emulating traffic at scale.
Raghu Nambiar, Corporate Vice President at AMD, highlighted the significance of this collaboration, stating that the ability of AMD EPYC CPUs to deliver parallel processing performance is crucial for handling the increasing volumes of data traffic in 5G networks.
Ram Periakaruppan, Vice President at Keysight, emphasized the inadequacy of single-dimension benchmarking methods and the necessity for new performance benchmarks in modern mobile and 5G networks. The use of AMD's 'Zen 4' chiplet architecture in this context demonstrates its capacity to meet complex network demands.
This partnership between Keysight and AMD represents a significant step in ensuring that data center operators can accurately provision their infrastructures to support the burgeoning growth of 5G and mobile network applications. The information for this article is based on a press release statement.
This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.