cert
cert-1
cert-2

Pass Cisco 300-640 Exam in First Attempt Guaranteed!

Get 100% Latest Exam Questions, Accurate & Verified Answers to Pass the Actual Exam!
30 Days Free Updates, Instant Download!

cert-5
cert-6
300-640 Exam - Verified By Experts
300-640 Premium File

300-640 Premium File

$89.99
$98.99
  • Premium File 60 Questions & Answers. Last Update: Nov 10, 2026

Whats Included:

  • Latest Questions
  • 100% Accurate Answers
  • Fast Exam Updates
 
$98.99
$89.99
accept 10 downloads in the last 7 days
block-screenshots
Cisco 300-640 Practice Test 14

Question 130

The acceptance board must decide whether an AI-infrastructure response to east-west training bandwidth is fit for controlled operation. The fictional operating facts are: An ad-ranking training farm shows strong single-node step times but cross-rack all-reduce performance collapses during job overlap. Signoff cannot rely on a good-looking utilization graph, a change ticket alone or an unmeasured promise. The response must include an appropriate governing engineering principle, a scoped intervention and a separate measurable acceptance criterion. Choose EXACTLY THREE statements that meet those roles.

  1. The multi-node training benchmark meets required epoch time without sustained fabric congestion.
  2. GPU selection must consider model parameters, activation memory, optimizer states, precision and parallelism rather than nominal FLOPS.
  3. Distributed training requires adequate effective east-west throughput between GPU workers, with oversubscription based on real collective traffic.
  4. Multiple tenants operate within their allocations while a full-device benchmark remains unaffected.
  5. Size and balance the leaf-spine fabric for the observed communication pattern before adding accelerator nodes.
  6. Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access.

Explanation

Correct Answer

A:

An ad-ranking training farm shows strong single-node step times but cross-rack all-reduce performance collapses during job overlap. The multi-node training benchmark meets required epoch time without sustained fabric congestion. Size and balance the leaf-spine fabric for the observed communication pattern before adding accelerator nodes.

C:

An ad-ranking training farm shows strong single-node step times but cross-rack all-reduce performance collapses during job overlap. Distributed training requires adequate effective east-west throughput between GPU workers, with oversubscription based on real collective traffic. Size and balance the leaf-spine fabric for the observed communication pattern before adding accelerator nodes.

E:

An ad-ranking training farm shows strong single-node step times but cross-rack all-reduce performance collapses during job overlap. Size and balance the leaf-spine fabric for the observed communication pattern before adding accelerator nodes. Distributed training requires adequate effective east-west throughput between GPU workers, with oversubscription based on real collective traffic.

Incorrect Answers

B:

GPU selection must consider model parameters, activation memory, optimizer states, precision and parallelism rather than nominal FLOPS. It is not the rule that supports the proposed change and acceptance test: An ad-ranking training farm shows strong single-node step times but cross-rack all-reduce performance collapses during job overlap. Distributed training requires adequate effective east-west throughput between GPU workers, with oversubscription based on real collective traffic. Size and balance the leaf-spine fabric for the observed communication pattern before adding accelerator nodes.

D:

Multiple tenants operate within their allocations while a full-device benchmark remains unaffected. It would not independently demonstrate that the control works: An ad-ranking training farm shows strong single-node step times but cross-rack all-reduce performance collapses during job overlap. The multi-node training benchmark meets required epoch time without sustained fabric congestion. Size and balance the leaf-spine fabric for the observed communication pattern before adding accelerator nodes.

F:

Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access. It does not implement the governing rule for this approval: An ad-ranking training farm shows strong single-node step times but cross-rack all-reduce performance collapses during job overlap. Size and balance the leaf-spine fabric for the observed communication pattern before adding accelerator nodes. The multi-node training benchmark meets required epoch time without sustained fabric congestion.

Question 131

A data-center AI architecture committee considers tail latency and network jitter in its AI Infrastructure Components and Architecture review. At the affected site: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. The committee is deliberately not selecting a vendor product or applying a live change yet. It needs the foundational reasoning that makes the next design decision defensible. Which ONE engineering principle directly governs this situation? Choose ONE answer.

  1. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.
  2. AI jobs split across sites need an explicit source-of-truth version and replication consistency contract.
  3. Serving capacity needs headroom for node failure, peak concurrency and warm model loading time.
  4. GPU partitioning and virtualization require platform-licensed, driver-compatible capabilities verified for the target workload.

Explanation

Correct Answer

A:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

Incorrect Answers

B:

AI jobs split across sites need an explicit source-of-truth version and replication consistency contract. It does not govern the condition described: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

C:

Serving capacity needs headroom for node failure, peak concurrency and warm model loading time. It does not govern the condition described: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

D:

GPU partitioning and virtualization require platform-licensed, driver-compatible capabilities verified for the target workload. It does not govern the condition described: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

Question 132

An infrastructure diagnostician is investigating tail latency and network jitter for the AI Infrastructure Components and Architecture team. The reported incident reads: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Competing teams attribute the symptoms to different layers of the AI stack. Before accepting a hypothesis or touching production settings, the investigator must choose a discriminating set of measurements. Which ONE evidence package is MOST relevant? Choose ONE answer.

  1. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.
  2. Model concurrent requests, cold-start time, per-replica memory, autoscaling delay and one-node-failed capacity.
  3. Inspect topology failure domains, ECMP next hops, routing timers and failover packet-loss measurements.
  4. Review rack nameplate ratings, measured peak draw, dual-feed loading, breaker headroom and PSU redundancy.

Explanation

Correct Answer

A:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

Incorrect Answers

B:

Model concurrent requests, cold-start time, per-replica memory, autoscaling delay and one-node-failed capacity. That check would not establish the decisive facts for this condition: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

C:

Inspect topology failure domains, ECMP next hops, routing timers and failover packet-loss measurements. That check would not establish the decisive facts for this condition: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

D:

Review rack nameplate ratings, measured peak draw, dual-feed loading, breaker headroom and PSU redundancy. That check would not establish the decisive facts for this condition: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

Question 133

During a bounded AI-platform change-control review concerning tail latency and network jitter, the site owner reports: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. The proposed work must have a narrow implementation scope, first be validated against the failure evidence and avoid unrelated system-wide tuning. Which ONE intervention should the engineers pilot before any wider rollout? Choose ONE answer.

  1. Compare like-for-like PUE baselines and optimize cooling or capacity with a documented energy measurement plan.
  2. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.
  3. Validate and stage a supported end-to-end software bill of materials with rollback images.
  4. Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access.

Explanation

Correct Answer

B:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

Incorrect Answers

A:

Compare like-for-like PUE baselines and optimize cooling or capacity with a documented energy measurement plan. That change would leave the reported problem unresolved: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

C:

Validate and stage a supported end-to-end software bill of materials with rollback images. That change would leave the reported problem unresolved: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

D:

Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access. That change would leave the reported problem unresolved: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

Question 134

Two independent technical reviewers examine a AI Infrastructure Components and Architecture problem involving tail latency and network jitter. Their shared incident log states: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. One reviewer requires measurements capable of separating the observed problem from neighboring failure modes; the other requires a bounded action that follows those measurements. Choose EXACTLY TWO statements that together supply those two parts.

  1. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.
  2. Use versioned immutable datasets with controlled transfer promotion and reject stale job inputs.
  3. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.
  4. Collect calibrated facility and IT power readings, interval timestamps, seasonal conditions and utilization context.
  5. Check firmware, drivers, hypervisor or container runtime compatibility, partition inventory and isolation tests.

Explanation

Correct Answer

A:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. Measured p99 inference response remains within budget during the approved peak-burst workload.

C:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Measured p99 inference response remains within budget during the approved peak-burst workload.

Incorrect Answers

B:

Use versioned immutable datasets with controlled transfer promotion and reject stale job inputs. It is not the scoped response supported by the accompanying evidence: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Measured p99 inference response remains within budget during the approved peak-burst workload.

D:

Collect calibrated facility and IT power readings, interval timestamps, seasonal conditions and utilization context. It would not provide the evidence needed for the paired diagnosis and follow-up: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. Measured p99 inference response remains within budget during the approved peak-burst workload.

E:

Check firmware, drivers, hypervisor or container runtime compatibility, partition inventory and isolation tests. It would not provide the evidence needed for the paired diagnosis and follow-up: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. Measured p99 inference response remains within budget during the approved peak-burst workload.

Question 135

The acceptance board must decide whether an AI-infrastructure response to tail latency and network jitter is fit for controlled operation. The fictional operating facts are: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Signoff cannot rely on a good-looking utilization graph, a change ticket alone or an unmeasured promise. The response must include an appropriate governing engineering principle, a scoped intervention and a separate measurable acceptance criterion. Choose EXACTLY THREE statements that meet those roles.

  1. GPU partitioning and virtualization require platform-licensed, driver-compatible capabilities verified for the target workload.
  2. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.
  3. Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access.
  4. At target worker count, training steps avoid input stalls and storage utilization remains bounded.
  5. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.
  6. Measured p99 inference response remains within budget during the approved peak-burst workload.

Explanation

Correct Answer

B:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

E:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

F:

An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Measured p99 inference response remains within budget during the approved peak-burst workload. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

Incorrect Answers

A:

GPU partitioning and virtualization require platform-licensed, driver-compatible capabilities verified for the target workload. It is not the rule that supports the proposed change and acceptance test: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

C:

Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access. It does not implement the governing rule for this approval: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Measured p99 inference response remains within budget during the approved peak-burst workload.

D:

At target worker count, training steps avoid input stalls and storage utilization remains bounded. It would not independently demonstrate that the control works: An insurance pricing service misses a 75-ms p99 target only during simultaneous model checkpoint transfers. Measured p99 inference response remains within budget during the approved peak-burst workload. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

Question 136

A data-center AI architecture committee considers tail latency and network jitter in its AI Infrastructure Components and Architecture review. At the affected site: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. The committee is deliberately not selecting a vendor product or applying a live change yet. It needs the foundational reasoning that makes the next design decision defensible. Which ONE engineering principle directly governs this situation? Choose ONE answer.

  1. GPU partitioning and virtualization require platform-licensed, driver-compatible capabilities verified for the target workload.
  2. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.
  3. Serving capacity needs headroom for node failure, peak concurrency and warm model loading time.
  4. Checkpoint systems must meet recovery-point and recovery-time needs with verified durable writes and restore validation.

Explanation

Correct Answer

B:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

Incorrect Answers

A:

GPU partitioning and virtualization require platform-licensed, driver-compatible capabilities verified for the target workload. It does not govern the condition described: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

C:

Serving capacity needs headroom for node failure, peak concurrency and warm model loading time. It does not govern the condition described: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

D:

Checkpoint systems must meet recovery-point and recovery-time needs with verified durable writes and restore validation. It does not govern the condition described: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

Question 137

An infrastructure diagnostician is investigating tail latency and network jitter for the AI Infrastructure Components and Architecture team. The reported incident reads: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Competing teams attribute the symptoms to different layers of the AI stack. Before accepting a hypothesis or touching production settings, the investigator must choose a discriminating set of measurements. Which ONE evidence package is MOST relevant? Choose ONE answer.

  1. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.
  2. Model concurrent requests, cold-start time, per-replica memory, autoscaling delay and one-node-failed capacity.
  3. Check firmware, drivers, hypervisor or container runtime compatibility, partition inventory and isolation tests.
  4. Count metadata operations, open latency, namespace contention and worker-level file listing times.

Explanation

Correct Answer

A:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

Incorrect Answers

B:

Model concurrent requests, cold-start time, per-replica memory, autoscaling delay and one-node-failed capacity. That check would not establish the decisive facts for this condition: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

C:

Check firmware, drivers, hypervisor or container runtime compatibility, partition inventory and isolation tests. That check would not establish the decisive facts for this condition: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

D:

Count metadata operations, open latency, namespace contention and worker-level file listing times. That check would not establish the decisive facts for this condition: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

Question 138

During a bounded AI-platform change-control review concerning tail latency and network jitter, the site owner reports: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. The proposed work must have a narrow implementation scope, first be validated against the failure evidence and avoid unrelated system-wide tuning. Which ONE intervention should the engineers pilot before any wider rollout? Choose ONE answer.

  1. Validate and stage a supported end-to-end software bill of materials with rollback images.
  2. Coordinate facility-validated cooling distribution and thermal controls before increasing sustained GPU power.
  3. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.
  4. Build physically independent eligible paths and test controlled link or leaf failure under active AI traffic.

Explanation

Correct Answer

C:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

Incorrect Answers

A:

Validate and stage a supported end-to-end software bill of materials with rollback images. That change would leave the reported problem unresolved: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

B:

Coordinate facility-validated cooling distribution and thermal controls before increasing sustained GPU power. That change would leave the reported problem unresolved: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

D:

Build physically independent eligible paths and test controlled link or leaf failure under active AI traffic. That change would leave the reported problem unresolved: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

Question 139

Two independent technical reviewers examine a AI Infrastructure Components and Architecture problem involving tail latency and network jitter. Their shared incident log states: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. One reviewer requires measurements capable of separating the observed problem from neighboring failure modes; the other requires a bounded action that follows those measurements. Choose EXACTLY TWO statements that together supply those two parts.

  1. Collect out-of-memory traces, per-rank allocator usage, peak activations, tensor sizes and chosen precision.
  2. Measure per-worker read concurrency, storage-node load, cache hit rates, queue depth and samples-per-second.
  3. Use workload-aware anti-affinity and recovery checkpoints across independent domains where supported.
  4. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.
  5. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load.

Explanation

Correct Answer

D:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Measured p99 inference response remains within budget during the approved peak-burst workload.

E:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. Measured p99 inference response remains within budget during the approved peak-burst workload.

Incorrect Answers

A:

Collect out-of-memory traces, per-rank allocator usage, peak activations, tensor sizes and chosen precision. It would not provide the evidence needed for the paired diagnosis and follow-up: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. Measured p99 inference response remains within budget during the approved peak-burst workload.

B:

Measure per-worker read concurrency, storage-node load, cache hit rates, queue depth and samples-per-second. It would not provide the evidence needed for the paired diagnosis and follow-up: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Correlate request p99 with per-hop queue occupancy, drops, ECN marks and burst timing rather than average link load. Measured p99 inference response remains within budget during the approved peak-burst workload.

C:

Use workload-aware anti-affinity and recovery checkpoints across independent domains where supported. It is not the scoped response supported by the accompanying evidence: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Measured p99 inference response remains within budget during the approved peak-burst workload.

Question 140

The acceptance board must decide whether an AI-infrastructure response to tail latency and network jitter is fit for controlled operation. The fictional operating facts are: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Signoff cannot rely on a good-looking utilization graph, a change ticket alone or an unmeasured promise. The response must include an appropriate governing engineering principle, a scoped intervention and a separate measurable acceptance criterion. Choose EXACTLY THREE statements that meet those roles.

  1. Measured p99 inference response remains within budget during the approved peak-burst workload.
  2. GPU selection must consider model parameters, activation memory, optimizer states, precision and parallelism rather than nominal FLOPS.
  3. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.
  4. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.
  5. Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access.
  6. The rack survives an authorized single-feed test without exceeding the remaining electrical capacity.

Explanation

Correct Answer

A:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Measured p99 inference response remains within budget during the approved peak-burst workload. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

C:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

D:

A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency.

Incorrect Answers

B:

GPU selection must consider model parameters, activation memory, optimizer states, precision and parallelism rather than nominal FLOPS. It is not the rule that supports the proposed change and acceptance test: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. A low-average-latency network can still fail serving SLOs when transient queuing creates high p99 latency. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

E:

Apply least-privilege segmentation and test allowed model/data flows plus blocked management and tenant access. It does not implement the governing rule for this approval: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay. Measured p99 inference response remains within budget during the approved peak-burst workload.

F:

The rack survives an authorized single-feed test without exceeding the remaining electrical capacity. It would not independently demonstrate that the control works: A customer-search inference tier reports 8-ms median RPC latency but occasional 250-ms network spikes under microbursts. Measured p99 inference response remains within budget during the approved peak-burst workload. Reserve appropriate service classes and relieve burst contention after validating which path contributes to tail delay.

Implementing Cisco Data Center AI Infrastructure (DCAI) (300-640) Practice Tests