Accelerate AI from edge to core with Cisco UCS X-Series
The new Cisco UCS X580p GPU node with UCS X-Fabric delivers GPU-dense performance, scalable fabrics, unified management, and supports NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs.
Cisco UCS X-Series redefines AI infrastructure flexibility. Mix GPU and CPU nodes in one modular chassis with X-Fabric for high-bandwidth, low-latency connectivity.
High GPU density for AI
UCS X580p supports up to four NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs and others, delivering massive parallel processing for LLM training, generative AI, and high-performance graphics workloads.
Flexible X-Fabric Interconnect
High-bandwidth, low-latency fabric enables PCIe Gen 5 switching, intra-host GPU Interconnect with NVLink, and inter-host scaling with RDMA over AI fabric.
Unified management with Cisco Intersight
Simplify operations with cloud-based lifecycle management, workload optimization, and AI-specific monitoring.
CPU-optimized blade for AI inference, analytics, and pre/post-processing.
Dual Xeon Scalable CPUs
Flexible memory/storage
X-Fabric enabled
Cisco RTX Pro Servers combine the modular power of UCS X-Series with NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. Designed for AI training, generative AI, and professional visualization, RTX Pro Servers deliver extreme performance with scalable fabrics and unified management.
X-Fabric: The performance backbone for AI
The UCS X-Series X-Fabric Interconnect enables dynamic resource allocation between GPU and CPU nodes, delivering high-speed, low-latency connectivity for data-intensive AI workloads—without re-architecting your infrastructure.
The UCS X580p node inside the modular UCS X-Series chassis supports up to four NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. As a Cisco RTX Pro Server, it delivers enterprise-class AI training, generative AI, and visualization performance with the flexibility of X-Fabric interconnects.
