350-501 Automation and Quality of Services Practice Test 2 — 15 Questions

Question 1

An SP engineer is configuring hierarchical QoS on a PE router for a subscriber with a 10 Mbps shaped rate. The child policy must guarantee 2 Mbps for voice, 4 Mbps for video, and the rest for data. However, the parent policy shape rate is 10 Mbps. Which configuration ensures proper operation?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Parent shape 10 Mbps, child policy with LLQ for voice and CBWFQ for video and data with bandwidth 5 Mbps each. The correct configuration must ensure that the child policy's bandwidth guarantees sum to no more than the parent shape rate (10 Mbps) and that voice gets priority treatment. Option A uses LLQ for voice (priority) and CBWFQ for video and data with bandwidth 5 Mbps each, totaling 10 Mbps, which fits within the parent shape. This provides guaranteed bandwidth for voice and video while allowing data to use the remaining capacity.

Answer

Parent shape 10 Mbps, child policy with priority for voice and bandwidth 4 Mbps for video, no data class

Answer

Parent shape 10 Mbps, child policy with bandwidth percent 20 voice, 40 video, and 40 data

Answer

Parent shape 10 Mbps, child policy with bandwidth 2 Mbps for voice, 4 Mbps for video, and default class for data

Question 2

Which DSCP value is recommended for Expedited Forwarding (EF) per-hop behavior, commonly used for voice traffic?

Accepted Answer

DSCP 46. DSCP 46 corresponds to the Expedited Forwarding (EF) per-hop behavior defined in RFC 3246, which is the standard marking for voice traffic requiring low latency, low jitter, and low loss. The decimal value 46 maps to binary 101110, and its upper three bits (101) place it in the IP Precedence 5 class historically used for voice.

Answer

DSCP 48

Answer

DSCP 34

Answer

DSCP 0

Question 3

An SP uses NSO to orchestrate MPLS L3VPN services. The service model is written in YANG, and the NED communicates with the PE routers. Which protocol does NSO use to push configuration to the network devices?

Accepted Answer

NETCONF. Cisco NSO uses NETCONF as the primary southbound protocol to push configuration to network devices, with YANG models defining the data schema. The NED (Network Element Driver) translates NSO's service model into device-specific NETCONF RPCs, typically over SSH on port 830.

Answer

CLI

Answer

PCEP

Answer

SNMP

Question 4

An SP is implementing traffic shaping on an MPLS PE-CE link. Which TWO statements are true about shaping compared to policing? (Select two.)

Accepted Answer

Policing can re-mark packets to a lower DSCP, shaping cannot.. Option D is correct because policing is typically implemented with a token bucket that can take actions such as re-marking out-of-profile packets to a lower DSCP (e.g., via MQC policy-map with 'police ... conform-action ... exceed-action set-dscp-transmit'), whereas shaping only delays packets and has no re-marking action. Option E is correct because shaping uses a token bucket with a queue/buffer to delay excess traffic so it conforms to the committed rate, while policing drops or re-marks packets that exceed the configured rate rather than buffering them. Option A is incorrect because shaping, not policing, introduces delay by buffering excess traffic; policing generally adds little delay. Option B is incorrect because both shaping and policing are typically applied on egress (outbound) interfaces, and shaping is not an ingress-only function. Option C is incorrect because shaping normally does not drop packets—it buffers and delays them—while policing may drop or re-mark, so neither always causes packet loss.

Answer

Policing typically introduces more delay than shaping.

Answer

Shaping is applied on ingress, policing on egress.

Answer

Both shaping and policing always result in packet loss.

Question 5

Which THREE DSCP values are part of the Assured Forwarding (AF) class 4? (Select three.)

Accepted Answer

DSCP 34. The Assured Forwarding (AF) class 4 consists of AF41, AF42, and AF43, whose DSCP values are 34, 36, and 38 respectively, so options A (DSCP 34), C (DSCP 36), and E (DSCP 38) are correct. AF4x is encoded as class 4 with drop precedence 1, 2, or 3, giving binary 100xx0 and decimal values 34, 36, and 38. Option B (DSCP 40) belongs to AF class 5 as AF41? No, DSCP 40 is AF41? Actually AF41 is 34, so 40 is CS5? DSCP 40 is CS5, not AF4. Option D (DSCP 46) is Expedited Forwarding (EF), not part of AF class 4. Therefore B and D are incorrect.

Answer

DSCP 40

Answer

DSCP 46

Question 6

An SP is using MQC to configure QoS on a core router. Which TWO statements about MQC are correct? (Select two.)

Accepted Answer

A class-map can match multiple criteria using 'match-all' or 'match-any'.. Option A is correct because MQC class-maps support logical operators 'match-all' (logical AND, the default) and 'match-any' (logical OR) to combine multiple match statements such as 'match protocol' or 'match dscp'. Option C is correct because every policy-map implicitly contains a 'class class-default' that handles traffic not matching any explicitly defined class, and it can be referenced explicitly to apply actions like policing or marking to unclassified traffic. Option B is incorrect because a policy-map is attached to interfaces, subinterfaces, or a control-plane/frame-relay/virtual-template via the 'service-policy' command, not applied globally to all interfaces at once. Option D is incorrect because 'service-policy' is issued in interface (or other attachment-point) configuration mode, not global configuration mode. Option E is incorrect because WRED is a congestion-avoidance mechanism configured inside a policy-map class via 'random-detect', so the statement that it can ONLY be configured there is a mischaracterization of its placement and scope.

Answer

A policy-map can be applied globally to all interfaces.

Answer

The 'service-policy' command is used in global configuration mode.

Answer

WRED can only be configured within a policy-map class.

Question 7

An SP is deploying SR-PCE for centralized traffic engineering. Which THREE functions does the PCE perform? (Select three.)

Accepted Answer

Computes path for traffic-engineered LSPs based on constraints.. Option A is correct because the PCE's core function is path computation: it runs CSPF-style algorithms over the TED to calculate constrained paths for TE LSPs, including SR-TE policies, based on attributes such as bandwidth, affinity, latency, and SRLG. Option C is correct because in an SR-PCE deployment the controller instantiates SR-TE policies on the headend routers (the PCCs), pushing the policy definition (color, endpoint, candidate paths, segment lists) so the routers can steer traffic into the SR path. Option D is correct because the PCE communicates computed paths to PCCs via PCEP, using PCRep/PCInitiate messages to return or initiate the segment list that the PCC installs. Option B is not a PCE function: BGP-LS is a topology-collection mechanism used by the PCE (or a BGP-LS speaker) to learn link-state information from IGPs, not something the PCE distributes. Option E is not a PCE function: installing MPLS forwarding entries (ILM/NHLFE) on routers is done locally by the routers themselves, not by the PCE.

Answer

Distributes link-state information using BGP-LS.

Answer

Configures MPLS forwarding on routers.

Question 8

An SP engineer is configuring QoS on a PE router for MPLS traffic. The core network uses MPLS EXP bits for QoS. Which TWO methods can be used to set MPLS EXP bits based on IP DSCP? (Select two.)

Accepted Answer

Use a policy-map with 'set mpls experimental' command.. Option C is correct because a policy-map with the 'set mpls experimental' command (or 'set mpls experimental topmost'/'imposition' depending on platform) explicitly maps classified IP DSCP values to MPLS EXP bits before or during label imposition. Option E is correct because 'set mpls experimental imposition' under a policy attached to the interface sets the EXP bits on the imposed label based on the matched DSCP/class, which is the standard MPLS QoS marking mechanism on Cisco PE routers. Options A, B, and D are incorrect: 'mpls qos enforce' is not a valid command for DSCP-to-EXP mapping, 'mpls ip' merely enables label switching and does not copy DSCP to EXP, and 'qos-group' is an internal classification marker that must still be mapped to EXP via a set command rather than directly setting EXP bits.

Answer

Configure 'mpls qos enforce' on the interface.

Answer

Configure 'mpls ip' to automatically copy DSCP to EXP.

Answer

Apply 'qos-group' to DSCP mapping.

Question 9

Which THREE of the following are standard per-hop behaviors (PHBs) in the DiffServ model? (Select three.)

Accepted Answer

Default (Best Effort). The DiffServ model defines EF (Expedited Forwarding), AF (Assured Forwarding), and Default (BE) PHBs. CS is Class Selector, but not a separate PHB; it is part of AF? Actually CS is a backward-compatible PHB. Standard ones are EF, AF, and Default.

Answer

LLQ

Answer

CS (Class Selector)

Question 10

A service provider is implementing a QoS policy using MQC (Modular QoS CLI) on Cisco IOS XR routers. The policy must classify traffic based on DSCP values and mark packets accordingly. The engineer needs to use a class map that matches multiple DSCP values efficiently. Which two statements are true about configuring class maps for QoS classification? (Choose two.)

Accepted Answer

A class map can match on a list of DSCP values using the 'match dscp' command with multiple values separated by spaces.. In Cisco IOS XR, a class map can match multiple DSCP values in a single 'match dscp' statement, and when multiple values are specified, the match is successful if any of the values match. This allows efficient classification of traffic that should receive the same QoS treatment. The other statements are false because they misrepresent the capabilities of class maps.

Answer

The 'match dscp' command can only match a single DSCP value per class map.

Answer

Class maps can only be used in policy maps for marking, not for policing or shaping.

Answer

Class maps used for QoS classification can match on both DSCP and IP precedence simultaneously in the same class map.

Question 11

In the DiffServ model, which DSCP value is used for Expedited Forwarding (EF) to support voice traffic?

Accepted Answer

DSCP 46. In the DiffServ model, Expedited Forwarding (EF) is defined by DSCP 46 (binary 101110), which provides low-latency, low-jitter, low-loss treatment suitable for voice traffic. EF is specified in RFC 3246 and is the standard marking for VoIP bearer traffic. DSCP 46 is the correct value for EF.

Answer

DSCP 0

Answer

DSCP 40

Answer

DSCP 34

Question 12

An SP network uses NSO for service orchestration. Which component in NSO provides the abstraction layer to communicate with different device types using protocols like NETCONF or CLI?

Accepted Answer

NED. NEDs (Network Element Drivers) are plugins that translate NSO's service models to device-specific configurations.

Answer

YANG

Answer

RESTCONF

Answer

CDB

Question 13

What is the primary difference between traffic policing and traffic shaping?

Accepted Answer

Shaping buffers excess traffic, while policing drops or re-marks it.. Traffic shaping buffers excess packets and delays them to smooth bursts to conform to a rate, while traffic policing drops or re-marks packets that exceed the configured rate. This is the fundamental behavioral difference between the two QoS mechanisms.

Answer

Policing buffers excess traffic, while shaping drops it.

Answer

Policing is only applied inbound, shaping only outbound.

Answer

Both policing and shaping buffer excess traffic.

Question 14

An SP engineer is implementing LLQ on a PE router to prioritize voice. Which additional configuration is recommended to prevent voice traffic from starving other queues?

Accepted Answer

Configure a policer on the LLQ class to limit its rate.. LLQ (Low Latency Queueing) gives priority queueing to voice traffic, which can starve other queues if not policed. Configuring a policer on the LLQ class limits the amount of traffic that can enter the priority queue, ensuring that voice does not consume all bandwidth and leaving capacity for other classes. This is a recommended best practice to prevent starvation.

Answer

Use CBWFQ without LLQ.

Answer

Disable queueing on other classes.

Answer

Set the priority queue bandwidth to 100%.

Question 15

In the context of YANG modeling for network automation, which statement about 'must' expressions is true?

Accepted Answer

They enforce a constraint that must be true for the data to be valid.. In YANG, a 'must' expression is a constraint that must evaluate to true for the data to be considered valid. It is used to enforce additional validation rules on data nodes, such as ensuring that one leaf's value is consistent with another. If the condition is false, the data is rejected.

Answer

They conditionally make a leaf mandatory.

Answer

They are used to refine types.

Answer

They are used to define default values.