Free Cisco 650-987 Exam Practice Questions & Explanations
Prepared & Reviewed by the ValidExamDumps Editorial Team
At ValidExamDumps, we consistently monitor updates to the Cisco 650-987 exam questions by Cisco. Whenever our team identifies changes in the exam questions, objectives, focus areas or requirements, We immediately update our exam questions for both PDF and online practice exams. This commitment ensures our customers always have access to the most current and accurate questions. By preparing with these up to date and 100% exam domain coverage questions, our customers can successfully pass the Cisco Data Center Unified Computing Design Specialist exam on their first attempt without needing additional materials or study guides.
Other certification materials providers often include outdated or removed questions by Cisco in their 650-987 exam. These outdated questions lead to customers failing their Cisco Data Center Unified Computing Design Specialist exam. In contrast, we ensure our questions bank includes only precise and up-to-date questions. Our main priority is your success in the Cisco 650-987 exam, not profiting from selling obsolete exam questions in PDF or Online Practice Test.
A designer proposes a UCS domain with two fabric interconnects in a cluster and states that the cluster provides active/active data forwarding for server traffic while UCS Manager itself runs as an active/standby (primary/subordinate) management plane.
Is this statement accurate?
True. In a UCS domain, both fabric interconnects forward data simultaneously (each blade's vNICs are pinned to fabric A or B, giving active/active data planes), while the UCS Manager management plane operates as a high-availability cluster with one primary and one subordinate FI, synchronized over the L1/L2 cluster links and using a shared virtual IP.
A customer plans to run Microsoft Windows Server and Red Hat Enterprise Linux bare-metal workloads on UCS B-Series blades booting from a Fibre Channel SAN over the UCS fabric. During design review, the customer asks what must be provided so the operating systems can properly use the VIC adapters and the multipathed SAN boot LUN.
Which two design actions are required? (Choose two.)
Bare-metal Windows and Linux require the Cisco eNIC (Ethernet) and fNIC (FC) drivers matched to the UCS firmware level per the hardware and software interoperability matrix; without fNIC the installer cannot see the SAN LUN. The boot policy must list SAN boot targets on both fabric A and B for redundancy, and the storage array must present the boot LUN as LUN 0 to the service profile's WWPNs.
Scrub policies erase disks and are unrelated; SAN boot means no local mirror is needed. Host firmware packages should be used to keep adapter firmware consistent, not disabled. Dynamic vNIC connection policies apply to VM-FEX with a hypervisor, not bare-metal OS installs.
A virtualization architect wants ESXi hosts on UCS B-Series blades to present each virtual machine's traffic to the fabric interconnect as an individual interface, so that per-VM policy, QoS and statistics are enforced in hardware and follow the VM during vMotion.
Which UCS/VMware design should be recommended?
VM-FEX uses dynamic vNICs on the Cisco VIC to create a virtual interface per VM on the fabric interconnect, delivering per-VM visibility, QoS and policy that follow the VM during vMotion (and optionally VMDirectPath hypervisor bypass).
A standard vSwitch or guest NIC teaming aggregates VM traffic behind host uplinks with no per-VM fabric visibility. Disjoint Layer 2 addresses upstream VLAN separation, not per-VM policy.
A design calls for Cisco UCS C-Series rack servers to be managed by UCS Manager with the same service profile model used for B-Series blades, while minimizing the number of cables and avoiding dedicated management cabling per server.
Which integration approach should the designer specify?
Single-wire (single-connect) management carries both management (CIMC) traffic and data traffic over the same VIC ports to a supported FEX or directly to the fabric interconnects, so the server is fully managed by UCS Manager with service profiles and minimal cabling.
Standalone Cisco IMC does not provide UCS Manager service profiles. Dual-wire management works but requires extra cables and a management FEX, which the requirement seeks to avoid. The fabric interconnect management port is for cluster/administrative access, not for server integration.
A customer is designing a UCS deployment for a VMware vSphere cluster of 64 blades. They require that any blade can be replaced by a spare within minutes, with the replacement server automatically assuming the same identity (UUID, MAC addresses, WWPNs) and boot-from-SAN configuration as the failed blade.
Which design element in Cisco UCS Manager best satisfies this requirement?
Stateless computing in UCS is achieved by abstracting identity into the service profile. Using updating service profile templates with UUID, MAC, WWNN/WWPN pools and a server pool lets the profile be re-associated (or automatically associated via pool qualification) to a spare blade, which then inherits the same identity and SAN boot targets.
Local disk mirroring protects data on a single server but does not move identity. BIOS policies standardize settings but carry no identity. The chassis discovery policy governs IOM-to-fabric-interconnect link behavior and is unrelated to server identity mobility.