Download Latest 300-640 Questions

In Cisco Intersight Managed Mode, Fabric Interconnect physical port roles and port-channel definitions are configured through the Port Policy . Cisco's current Intersight configuration documentation states that a Port Policy configures ports and port roles on a Fabric Interconnect and permits administrators to enable or disable individual ports or port channels. Supported roles include Ethernet uplinks and Ethernet uplink port channels.

The Port Policy is subsequently associated with the UCS Domain Profile. Cisco documents that a Domain Profile configures an FI pair through reusable policies and controls ports, port channels, VLANs, and VSANs. During Domain Profile creation, the administrator attaches the appropriate Port Policy to each Fabric Interconnect.

A LAN Connectivity Policy is used primarily at the server-profile level to define vNIC connectivity. A Network Connectivity Policy configures DNS-related server-management settings rather than FI uplink topology. An Ethernet Network Group Policy defines VLAN groups associated with Ethernet connectivity and can be referenced by uplink configurations, but it does not itself create the FI port channel.

Consequently, the policy that must be changed to build the multiple-100-GbE uplink port channel is the Port Policy .

Study Guide Reference: AI Infrastructure Deployment and Data Management — Intersight Managed Mode, UCS Domain Profiles, Fabric Interconnect port configuration, and uplink resiliency.

===============

Congestion Notification Packets are control-feedback messages used by RoCEv2 congestion-management mechanisms. When a receiving endpoint observes an ECN-marked packet, it generates a CNP toward the sender, instructing the transmitting endpoint to reduce its transmission rate. Losing these packets would delay congestion response and allow queues to continue building.

Cisco's validated RoCEv2-over-VXLAN design explicitly states that CNP packets use DSCP 48 and are classified into a high-priority queue . In Cisco's example, ordinary RoCEv2 traffic is mapped to QoS group 3, while CNP is mapped to QoS group 7. Cisco further states that the high-priority classification ensures that CNP packets are not dropped in the network.

Traffic shaping in option A would regulate CNP transmission rather than guarantee timely feedback. Option C places CNP in the same potentially congested queue as the data traffic that generated the congestion, defeating the purpose of reliable feedback. Option D is also incorrect because PFC is intended to protect a lossless data priority; CNP itself should be forwarded through a high-priority queue and should not depend on pause behavior.

Study Guide Reference: AI Infrastructure Deployment and Data Management — RoCEv2, ECN, CNP, QoS classification, and congestion management.

===============

Cisco UCS X-Series is the strongest fit because it is designed as a modular, scalable compute platform that can support conventional enterprise workloads and GPU-accelerated AI workloads within the same architectural framework. Cisco explicitly positions UCS X-Series for workload consolidation, allowing compute nodes and GPU resources to be added as requirements evolve without replacing the entire infrastructure. The platform supports substantial memory capacity, virtualization, GPU acceleration, and centralized lifecycle management through Cisco Intersight.

The current UCS X-Series portfolio supports X210c and X215c compute nodes, X580p GPU expansion, and X-Fabric connectivity. Cisco further positions the platform for AI workloads through high-bandwidth, low-latency GPU connectivity and scalable GPU and CPU configurations.

UCS M-Series is not the current strategic modular platform for these requirements. Unified Edge is optimized for distributed edge deployments rather than a large, expanding enterprise data center. A standalone C-Series deployment can provide powerful GPU servers, but it does not offer the same chassis-level composability and resource expansion model described in the scenario.

Study Guide Reference: 2.0 AI Infrastructure Components and Architecture — 2.2 Evaluate compute deployment based on CPU, GPU, memory, virtualization, scalability, redundancy, and workload requirements.

===============

An IP-based shared storage array is the appropriate architecture for a RAG environment containing approximately 50 TB of enterprise documents. RAG requires compute nodes to access a common body of source documents, embeddings, vector indexes, model artifacts, and intermediate datasets. With six UCS X210c servers participating in the solution, centralized shared storage avoids creating isolated data silos on individual compute nodes and allows the full AI cluster to access the same authoritative dataset.

Cisco AI POD and RAG architectures use enterprise shared storage connected through high-speed Ethernet. Cisco validates IP-based protocols including NFS, iSCSI, NVMe/TCP, and—in higher-performance AI platforms—RDMA-capable storage designs. Cisco's RAG architecture with FlexPod, for example, uses NFS, iSCSI, and NVMe/TCP storage networks, while Cisco AI PODs support external storage from partners such as NetApp, Pure Storage, and VAST Data.

JBOD and RAID 5 local storage remain tied to individual servers and do not provide the desired shared data layer. Tape provides economical archival retention but lacks the random-access latency and throughput required for active RAG retrieval.

Study Guide Reference: AI Infrastructure Components and Architecture — storage selection, AI data access, scalability, RAG infrastructure, and shared storage.

===============