Beijing, June 8, 2026 — At COMPUTEX 2026, held in Taipei from June 2 to 5, Xinertel introduced three core testing solutions for the Asia-Pacific market, covering AI computing networks, lossless networking, and embodied intelligence. Designed to support the entire lifecycle from R&D validation and mass production to real-world deployment, the solutions address key challenges emerging during AI industry development, including unstable AI cluster networking, high-speed device interoperability, and the lack of comprehensive testing systems for intelligent robotics.
A major highlight at the exhibition was Xinertel's high-speed AI computing network testing solution, built on the company's self-developed DarYu-X high-end testing platform and Renix software. Supporting 400G, 800G, and 1.6T testing, the solution enables end-to-end validation of optical modules, AI switches, DPUs, and GPU clusters. It helps equipment manufacturers and AI infrastructure providers perform interoperability verification, long-duration stability testing, and distributed AI training traffic simulation while accurately evaluating key performance indicators such as throughput and latency. The solution supports both product development and automated production testing, helping customers improve validation efficiency and accelerate deployment of AI computing infrastructure.
Xinertel also showcased its lossless network testing solution, designed to meet the low-latency and zero-packet-loss requirements of AI data centers. Based on key lossless Ethernet technologies including RoCEv2, Priority Flow Control (PFC), and Explicit Congestion Notification (ECN), the solution supports standardized network validation and realistic simulation of AI collective communication traffic. It enables testing of communication patterns such as AllReduce, AllGather, Broadcast, and ReduceScatter, while simulating concurrent AI training workloads with different model architectures, parallel strategies, and communication modes. Combined with mixed-priority traffic generation, network impairment injection, and leaf-spine topology simulation, the solution helps customers identify congestion, traffic fluctuations, and transmission interruptions while validating the stability of large-scale AI computing networks.
Xinertel further presented its integrated embodied intelligence testing solution, targeting the rapidly growing humanoid robotics and collaborative robotics industries. Covering both external network communication and internal industrial bus testing, the solution supports comprehensive validation of network performance, authentication, encryption, security resilience, and network impairment scenarios. It can simulate large-scale intelligent agent communication, evaluate latency, packet loss, and throughput, perform EtherCAT master-slave communication testing, and reproduce cybersecurity threats such as DDoS attacks and prompt injection. Xinertel is also expanding its testing capabilities to emerging technologies such as TSN convergence, helping customers ensure stable, reliable, and secure communications for next-generation intelligent systems.
Throughout the exhibition, Xinertel's self-developed technologies and mature testing solutions attracted strong interest from industry visitors. Leading ODM manufacturers, cloud data center service providers, AI infrastructure integrators, and intelligent robotics developers from Taiwan, together with customers and industry professionals from Japan, South Korea, and Southeast Asia, visited the booth for technical discussions. Xinertel's engineering team provided live demonstrations and solution consultations, working closely with customers to address practical challenges in product development, production deployment, and AI infrastructure implementation.
COMPUTEX 2026 provided Xinertel with an important platform to strengthen collaboration across the AI industry ecosystem. Through in-depth technical exchanges with dozens of key customers, the company established multiple opportunities for sample evaluation, POC(proof-of-concept) projects, and strategic cooperation, further demonstrating its growing technical influence and market competitiveness across the Asia-Pacific AI industry.
Previous: None
