Cisco DCCOR 350-601 (Data Center Core) Practice Exam

What the Cisco DCCOR 350-601 (Data Center Core) exam covers

• Network191 questions
• Compute191 questions
• Storage Network152 questions
• Automation and Artificial Intelligence114 questions
• Security115 questions

Free Cisco DCCOR 350-601 (Data Center Core) sample questions

A sample of 10 questions with answers and explanations. Sign up free to practice all 763.

1. Question 1Network

   Which OSPF area type suits a data centre leaf that must reach external routes but carries no external LSAs?

   • AA stub area, which blocks external LSAs and uses a default route insteadCorrect
   • BA backbone area, which carries every LSA type without restriction
   • CA not so stubby area, which permits external routes to be originated within it
   • DA transit area, which carries traffic for a virtual link between areas
   ✓ Correct answer: A

   Blocking external advertisements and injecting a default keeps the leaf's link state database small, which matters when the same design is repeated across many switches. The backbone carries everything, a not so stubby area exists to originate external routes locally, and a transit area carries virtual link traffic.

   Why the other options are wrong

   • BThe backbone area carries every LSA type rather than restricting them.
   • CA not so stubby area exists so external routes can be originated inside it.
   • DA transit area carries traffic for a virtual link rather than limiting LSAs.
2. Question 2Network

   Which is the first check when a newly built bundle does not come up?

   • AWhether both ends agree on the negotiation mode and the member configurationCorrect
   • BWhether the switches are running the same software release
   • CWhether the cabling uses the same manufacturer at both ends
   • DWhether the switches have the same number of installed line cards
   ✓ Correct answer: A

   Two passive ends and a mismatched member parameter account for the great majority of bundles that never come up, and both are visible within seconds of looking. Matching software releases, cable manufacturers and installed line card counts are three factors that almost never prevent a bundle forming.

   Why the other options are wrong

   • BDifferent software releases rarely prevent a bundle forming.
   • CThe cable manufacturer does not affect bundle negotiation.
   • DThe number of installed line cards is irrelevant to a bundle.
3. Question 3NetworkSelect all that apply

   Which TWO benefits does a fabric wide monitoring platform provide over per switch tools?

   • AEvents from several switches are correlated into one view of a single incidentCorrect
   • BHistorical data is retained beyond what an individual switch can hold locallyCorrect
   • CEach switch no longer needs a management address of any kind on the network
   • DEach switch no longer needs its configuration to be backed up before a change
   ✓ Correct answer: A, B

   The pair is what turns a set of unrelated alerts into a description of one fault, and it is what makes it possible to answer a question about last Tuesday at all. Removing the need for management addressing and removing the need for configuration backups are two claims that do not follow.

   Why the other options are wrong

   • CEach switch still needs a management address to be reached and monitored.
   • DConfiguration backups before a change remain necessary regardless of monitoring.
4. Question 4Compute

   Which describes a modular server chassis that presents compute nodes to a shared fabric?

   • ACompute nodes share the chassis power, cooling and fabric connectivityCorrect
   • BCompute nodes each carry their own power supply and cooling independently
   • CCompute nodes must all run the same operating system release as each other
   • DCompute nodes share their memory so the chassis appears as one large server
   ✓ Correct answer: A

   Sharing the infrastructure and not the compute is the whole design, which is why a chassis raises density without making one failure affect every node in it. Independent power per node, a shared operating system requirement and shared memory are three other arrangements.

   Why the other options are wrong

   • BCompute nodes do not each carry their own power supply and cooling.
   • CCompute nodes are not required to run the same operating system release.
   • DCompute nodes do not share memory to appear as one large server.
5. Question 5Compute

   Which is the practical reason for keeping headroom in a hyperconverged cluster?

   • AA failure must leave enough capacity to hold the rebuilt copies and keep servingCorrect
   • BA failure must leave enough licences to cover the nodes that are still running
   • CA failure must leave enough rack space to install a replacement node quickly
   • DA failure must leave enough operators available to attend the data centre
   ✓ Correct answer: A

   A cluster full to the brim cannot restore its own resilience, so it stays exposed until somebody adds hardware, which is the opposite of what the design promised. Licences, rack space and operator availability are three real constraints that this particular headroom is not about.

   Why the other options are wrong

   • BLicences for the running nodes are a commercial matter rather than capacity.
   • CRack space for a replacement node is a facilities matter rather than capacity.
   • DHow many operators are available to attend is a staffing matter, not capacity.
6. Question 6Compute

   Which describes the effect of automating a process that was badly designed by hand?

   • AThe same mistakes are now made faster and more consistently across the estateCorrect
   • BThe mistakes are corrected automatically because the tool applies best practice
   • CThe process becomes slower because the tool adds overhead to each step
   • DThe process becomes safer because a tool cannot make the errors a person makes
   ✓ Correct answer: A

   Automation is an amplifier rather than a corrective, which is why the design is worth fixing before it is encoded rather than afterwards. Automatic correction, added slowness and inherent safety are three comfortable assumptions that regularly turn out to be false.

   Why the other options are wrong

   • BThe tool applies what it was told rather than correcting the design.
   • CThe process does not become slower merely because a tool is running it.
   • DA tool makes different errors rather than none, so safety is not inherent.
7. Question 7Storage Network

   Which describes a slow drain device in a storage fabric?

   • AA device that accepts frames slower than they are sent, holding up the fabricCorrect
   • BA device that sends frames faster than the fabric is able to forward them
   • CA device that has stopped responding entirely and been removed from the fabric
   • DA device that has been zoned incorrectly and receives frames meant for another
   ✓ Correct answer: A

   The damage spreads well beyond the slow device, because the frames waiting for it occupy buffers that other conversations need, which is why one struggling host can degrade an entire fabric. A fast sender, a dead device and a mis zoned device are three different conditions.

   Why the other options are wrong

   • BA device sending faster than the fabric forwards is a different condition.
   • CA device that has stopped responding entirely is simply offline.
   • DA device receiving frames meant for another is a zoning error.
8. Question 8Automation and Artificial Intelligence

   Which describes a data model that defines the structure a device's configuration must follow?

   • AA schema stating which nodes exist, their types and how they nestCorrect
   • BA transcript of the commands an operator typed during configuration
   • CA picture of the network topology showing how the devices connect
   • DA log of the configuration changes applied to the device over time
   ✓ Correct answer: A

   Having the structure defined separately from any one protocol is what allows the same configuration to be carried several ways and validated before it is sent. A command transcript, a topology picture and a change log are three other artefacts that describe a device without defining its shape.

   Why the other options are wrong

   • BA transcript of typed commands records actions rather than defining structure.
   • CA picture of the topology shows connections rather than configuration shape.
   • DA log of changes over time records history rather than structure.
9. Question 9Automation and Artificial Intelligence

   Which is the practical consequence of automating against a controller instead of each device?

   • AFar less code is needed, and the controller resolves how each device is programmedCorrect
   • BFar more code is needed, because the controller adds a layer to work through
   • CThe same amount of code is needed, because the work simply moves elsewhere
   • DNo code is needed at all, because the controller configures itself entirely
   ✓ Correct answer: A

   Expressing what you want rather than how each device achieves it is the difference, and it also means the automation survives a device being replaced with a different model. More code, an identical amount and no code at all are three claims that do not follow.

   Why the other options are wrong

   • BThe controller removes work rather than adding a layer to work through.
   • CThe amount of code genuinely falls rather than simply moving.
   • DThe controller does not configure itself without being told the intent.
10. Question 10Security

    Which describes an alert only mode for a segmentation policy?

    • ATraffic that would be blocked is reported and still permitted throughCorrect
    • BTraffic that would be blocked is dropped and the drop is reported
    • CTraffic is blocked silently and no report is produced at all
    • DTraffic is permitted and no report of any kind is generated
    ✓ Correct answer: A

    It is much the safest way to find the flows that the observation period managed to miss, because the cost of an incomplete policy becomes a report to read rather than an outage to explain. Dropping the traffic and reporting the drop, blocking silently with no report and permitting with no report are three other modes.

    Why the other options are wrong

    • BDropping the traffic and reporting it is full enforcement with logging.
    • CBlocking silently gives no information about what was affected.
    • DPermitting without any report tells you nothing about the policy.

Who this Cisco DCCOR 350-601 (Data Center Core) practice exam is for

This practice set is for anyone preparing for the Cisco DCCOR 350-601 (Data Center Core) exam - from first-time candidates building a foundation to experienced Cisco practitioners doing a final review before test day. If you learn best by working through realistic questions and reading why each answer is right or wrong, it is built for you.

How to use this Cisco DCCOR 350-601 (Data Center Core) practice exam

1. Start with the free sample questions above to gauge your current baseline.
2. Read the full explanation on every question, including why each wrong option is wrong.
3. Track your weak domains and focus your study where you are losing the most marks.
4. Once you are scoring consistently well, take a timed, full-length mock exam.
5. Use your readiness score to decide when you are ready to book the real Cisco DCCOR 350-601 (Data Center Core) exam.

Related Cisco resources

• Cisco DCCOR 350-601 (Data Center Core) study guideKey concepts
• Cisco practice examsAll Cisco
• Certification pathWhere this fits
• Certification exam guides & tipsBlog
• Plans & pricingFree & paid
• Cisco CCNP Security SCOR (350-701) practice examRelated
• Cisco CCST Cybersecurity (100-160) practice examRelated
• Cisco CBRCOR 350-201 (Cybersecurity Professional) practice examRelated

Cisco DCCOR 350-601 (Data Center Core) practice exam FAQ