21 Models, Media & Topologies Practice Questions & Answers

1. 1. An installer must run 10GBASE-T links with a channel length of about 90 meters. Which statement about balanced twisted-pair cabling categories correctly explains the required choice?

   • A.Category 5e supports 10GBASE-T to 100 meters as long as the cable is shielded.
   • B.Category 6 supports 10GBASE-T only to roughly 55 meters, so Category 6a is needed for a full 100-meter channel.✓ Answer
   • C.Category 8 is the only category ever rated for 10GBASE-T, at any length.
   • D.Category 6 supports 10GBASE-T over the full 100-meter channel, so Category 6a adds nothing.

   10GBASE-T over Category 6 is limited to roughly 55 meters because of alien crosstalk between adjacent cables; Category 6a is specified with tighter alien-crosstalk performance and is tested to 500 MHz, which is what allows 10GBASE-T over the full 100-meter channel. Category 5e is not qualified for 10GBASE-T, and Category 8 is short-reach data center cabling limited to about 30 meters.

   Source: IEEE 802.3an 10GBASE-T; TIA-568 commercial building telecommunications cabling standardReport a problem with this question

2. 2. Why does single-mode fiber support far longer link distances than multimode fiber at the same data rate?

   • A.Its core of roughly 8 to 9 microns allows essentially one propagation path, so modal dispersion is eliminated.✓ Answer
   • B.Its larger core admits more light, which makes the received signal stronger.
   • C.It uses an LED source, and LEDs attenuate less than lasers.
   • D.It is metallically shielded, so it resists electromagnetic interference that multimode cannot.

   Distance on multimode is limited mainly by modal dispersion: the 50 or 62.5 micron core lets light travel along many paths of different lengths, so pulses spread and eventually overlap. A single-mode core is small enough that essentially one mode propagates, so a narrow laser source at 1310 or 1550 nm can be carried for many kilometers. Neither fiber type conducts electricity, so both are immune to electromagnetic interference.

   Source: TIA-568 commercial building telecommunications cabling standard — single-mode and multimode optical fiberReport a problem with this question

3. 3. A 40 Gigabit Ethernet link uses parallel optics, sending several lanes over a ribbon of fibers terminated in a single ferrule. Which connector type is designed for that multi-fiber termination?

   • A.ST, a round bayonet connector holding one fiber per ferrule.
   • B.MPO, a multi-fiber connector that holds 12 or 24 fibers in one ferrule.✓ Answer
   • C.LC, a small form-factor connector normally used as a single-fiber duplex pair.
   • D.SC, a square push-pull connector holding one fiber per ferrule.

   Parallel optics such as 40GBASE-SR4 use four transmit and four receive lanes at the same time, so all of the fibers must be terminated together; the MPO ferrule carries 12 or 24 fibers and can be broken out into duplex LC pairs. LC, SC and ST ferrules each hold a single fiber, so a duplex link built from them uses two ferrules.

   Source: TIA-568 commercial building telecommunications cabling standard — MPO multi-fiber connectors; IEEE 802.3 40GBASE-SR4Report a problem with this question

4. 4. Cable must be pulled through the space above a suspended ceiling that is used to return air to the building's air handling system. Why is plenum-rated (CMP) cable required there?

   • A.Plenum-rated cable carries higher data rates than riser-rated cable of the same category.
   • B.Its jacket uses material that resists flame spread and gives off less toxic smoke, which is required in a space that circulates building air.✓ Answer
   • C.The plenum jacket shields the pairs from interference produced by the air handling equipment.
   • D.Only plenum-rated cable is allowed to exceed a 100-meter run.

   The plenum listing is a fire-safety property of the jacket, not a transmission property: in an air-handling space, an ordinary PVC jacket that burns would push smoke through the whole ventilation system, so a low-smoke, flame-retardant jacket is mandated. Riser-rated cable is the equivalent listing for vertical shafts between floors, and a plenum cable of a given category performs exactly like a non-plenum cable of that category.

   Source: National Electrical Code communications cable listings (CMP plenum and CMR riser ratings)Report a problem with this question

5. 5. In an Ethernet designation such as 1000BASE-LX, what does the 'BASE' portion of the name indicate?

   • A.That the medium is copper rather than fiber.
   • B.That the line code encodes data in base 2.
   • C.That the maximum segment length is the base value of 100 meters.
   • D.That the medium carries a single signal using its entire bandwidth, which is baseband signaling.✓ Answer

   Ethernet names encode three things: the leading number is the signaling rate in megabits per second (1000 means 1000 Mbps), 'BASE' means baseband — the whole bandwidth of the medium carries one signal, as opposed to broadband, where the medium is split into frequency channels — and the suffix identifies the medium or physical layer, so LX is a long-wavelength laser over fiber while T is twisted pair.

   Source: IEEE 802.3 Ethernet nomenclature (baseband signaling)Report a problem with this question

6. 6. Two switches in the same rack must be connected at 10 Gbps over about two meters at the lowest cost. Which option describes a fixed-length twinaxial copper assembly with the transceivers permanently attached at both ends?

   • A.A Category 6a patch cord with RJ45 plugs.
   • B.A single-mode fiber jumper with SC connectors plus long-reach optics.
   • C.A direct attach copper (DAC) cable.✓ Answer
   • D.A multimode fiber patch cord with LC connectors plus a pair of SFP+ optics.

   A DAC is a twinaxial copper cable sold in fixed lengths with the transceiver housings attached to each end, so there are no separate optics to buy and no fiber end faces to clean; it is the standard inexpensive intra-rack link at 10 Gbps and above, but it cannot be re-terminated and reaches only a few meters. The other choices all require separate transceivers or a different medium.

   Source: IEEE 802.3 direct attach copper (twinaxial) cable assemblies and pluggable transceiver form factorsReport a problem with this question

7. 7. An access switch does not support Power over Ethernet, so a power injector is installed in the cable run between that switch port and a new wireless access point. What is this method of delivering power called?

   • A.Midspan delivery, because the power sourcing equipment sits between the switch and the powered device.✓ Answer
   • B.Endspan delivery, because the power is added at the end of the cable run.
   • C.It has no name, because IEEE Power over Ethernet works only from a PoE-capable switch.
   • D.Passive delivery, which is the method defined by IEEE 802.3af.

   IEEE Power over Ethernet defines two placements for the power sourcing equipment: an endspan PSE is the switch itself, energizing the pairs on its own ports, while a midspan PSE is a separate injector inserted partway along the link. Both negotiate with the powered device through the same detection and classification process, and the amendments raise available power in steps — roughly 15.4 W at the PSE for 802.3af, about 30 W for 802.3at, and up to about 90 W for 802.3bt using all four pairs.

   Source: IEEE 802.3af, 802.3at and 802.3bt Power over Ethernet — endspan and midspan power sourcing equipmentReport a problem with this question

8. 8. In the 2.4 GHz band the 802.11 channel numbers are spaced 5 MHz apart, while a standard channel occupies about 20 MHz of spectrum. What does that arrangement mean for channel planning?

   • A.Overlap disappears if both access points increase their transmit power.
   • B.Any two different channel numbers can be used on neighboring access points without interfering.
   • C.Two channels must differ by at least five channel numbers before their occupied spectrum stops overlapping.✓ Answer
   • D.Overlap matters only when the two access points advertise different network names.

   Because each carrier is about 20 MHz wide while the numbering grid steps only 5 MHz, a channel spills across roughly four neighboring channel numbers on each side; only channels separated by five or more numbers occupy disjoint spectrum. Overlapping channels force the radios to treat each other as noise rather than as traffic they can defer to, which is why neighboring cells are assigned non-overlapping channels, and raising power makes co-channel and adjacent-channel interference worse rather than better.

   Source: IEEE 802.11 channel plan for the 2.4 GHz band (5 MHz channel spacing, 20 MHz channel width)Report a problem with this question

9. 9. An engineer deploys identical dual-band access points at company offices in several different countries. What determines which channels and transmit power levels each access point is allowed to use?

   • A.The regulatory domain in which the access point is installed, since radio rules differ by country.✓ Answer
   • B.The 802.11 amendment that the access point hardware supports.
   • C.The number of clients associated with each radio.
   • D.The wireless network name configured on the radio.

   IEEE 802.11 defines the modulation, the channel widths and the channel center frequencies, but each national regulator decides which of those channels are licence-exempt and how much power may be radiated. Access points therefore carry a regulatory-domain setting, advertise country information to clients, and in the affected 5 GHz sub-bands must also apply dynamic frequency selection and transmit power control so that radar systems are protected.

   Source: IEEE 802.11 regulatory domain requirements, including 802.11d country information and 802.11h dynamic frequency selection and transmit power controlReport a problem with this question

10. 10. An administrator bonds two adjacent 20 MHz channels into a single 40 MHz channel on an access point. What is the expected trade-off?

    • A.Both throughput and the number of available channels rise, because bonding creates additional spectrum.
    • B.Peak throughput on that radio rises, but fewer non-overlapping channels remain for neighboring access points.✓ Answer
    • C.Range increases, because a wider channel concentrates more power in each hertz.
    • D.Throughput is unchanged, but the wider channel becomes immune to interference.

    Bonding roughly doubles the number of usable subcarriers, so the data rate roughly doubles, but the band contains a fixed amount of spectrum: every bonded channel consumes two channels' worth, halving how many non-overlapping cells can be planned. That is why bonding is practical in the wider 5 GHz and 6 GHz bands and is avoided in 2.4 GHz, where one bonded channel swallows most of the usable spectrum. Spreading the same transmit power over twice the bandwidth also lowers, not raises, power spectral density and range.

    Source: IEEE 802.11n channel bonding (20 MHz and 40 MHz operation)Report a problem with this question

11. 11. Which statement describes a durable characteristic of the IEEE 802.11n amendment?

    • A.It operates only in the 5 GHz band.
    • B.It introduced multiple input multiple output with several spatial streams and can operate in both the 2.4 GHz and 5 GHz bands.✓ Answer
    • C.It operates only in the 2.4 GHz band and is fixed at a 20 MHz channel width.
    • D.It was the first amendment to add downlink multi-user MIMO.

    The defining feature of 802.11n is MIMO: multiple antennas carry independent spatial streams at the same time, multiplying throughput, and the amendment also added optional 40 MHz channels; it is specified for both the 2.4 GHz and 5 GHz bands. By contrast 802.11a is 5 GHz only and 802.11g is 2.4 GHz only, while downlink multi-user MIMO — serving several clients simultaneously — first appeared in 802.11ac, which operates in 5 GHz.

    Source: IEEE 802.11n — MIMO and dual-band operationReport a problem with this question

12. 12. A wireless bridge must link two buildings roughly 500 meters apart with clear line of sight. Which antenna choice fits, and why?

    • A.An omnidirectional antenna at each end, because free-space path loss affects only directional antennas.
    • B.An omnidirectional dipole at each end, because a 360-degree pattern guarantees the other building is covered.
    • C.Either type performs the same, because free-space path loss does not depend on distance.
    • D.A directional antenna such as a parabolic dish or Yagi at each end, because focusing the radiation pattern into a narrow beam raises gain along the path and offsets free-space path loss.✓ Answer

    Signal strength falls with the square of distance in free space, so a long outdoor hop needs gain rather than area coverage. A directional antenna does not create power; it reshapes the radiation pattern, trading the wide pattern of a dipole for a narrow beam of much higher gain aimed at the far end, which is exactly what a point-to-point bridge needs. Omnidirectional antennas are meant to cover an area around the antenna, such as an open office floor.

    Source: CompTIA Network+ — Network Implementation domain (antenna types and radio propagation)Report a problem with this question

13. 13. Several access points in one building all advertise the wireless network name CORP so that clients can roam between them. What identifies one individual access point radio inside that network?

    • A.The channel width configured on that radio.
    • B.The SSID, which is the network name itself.
    • C.The ESSID, which is the name shared by every access point in the set.
    • D.The BSSID, which is the MAC address of that individual radio.✓ Answer

    The SSID is the human-readable network name; when the same SSID is shared by several access points to form one extended service set it is commonly called the ESSID, and that shared name is what lets a client roam. Each individual radio still identifies its own basic service set with a BSSID, which is a MAC address, so a dual-band access point presents different BSSIDs on its 2.4 GHz and 5 GHz radios while advertising the same SSID.

    Source: IEEE 802.11 service set identifiers — SSID, BSSID and extended service setReport a problem with this question

14. 14. As data from a TCP-based application is passed down the stack, what is the protocol data unit called at the Transport layer, and what is it called at the Network layer immediately below it?

    • A.Packet at the Transport layer and frame at the Network layer.
    • B.Segment at the Transport layer and packet at the Network layer.✓ Answer
    • C.Frame at the Transport layer and packet at the Network layer.
    • D.Datagram at the Transport layer and frame at the Network layer.

    Each layer adds its own header and renames the unit it hands down: TCP produces segments (the UDP equivalent is called a datagram), the Network layer encapsulates a segment into a packet carrying IP addressing, the Data Link layer wraps that into a frame carrying MAC addressing, and the Physical layer transmits the result as bits or symbols. On receipt the headers are stripped in the reverse order.

    Source: ISO/IEC 7498 OSI Basic Reference Model — protocol data unitsReport a problem with this question

15. 15. The OSI model defines separate Session, Presentation and Application layers. How are those three sets of functions represented in the four-layer TCP/IP model?

    • A.They are collapsed into the TCP/IP Transport layer.
    • B.The TCP/IP model has no equivalent for any of them.
    • C.Session maps to Transport, while Presentation and Application map to the Internet layer.
    • D.They are collapsed into the single TCP/IP Application layer.✓ Answer

    The TCP/IP model has four layers — Link, Internet, Transport and Application — and everything OSI splits across layers five, six and seven is handled inside the TCP/IP Application layer, usually by the application protocol itself or its libraries. The rest of the mapping is straightforward: OSI Physical and Data Link correspond to the Link layer, OSI Network to the Internet layer, and OSI Transport to Transport.

    Source: RFC 1122 Requirements for Internet Hosts — TCP/IP layeringReport a problem with this question

16. 16. An eight-port switch has a single host connected to each port, and every port belongs to the same VLAN. How many collision domains and how many broadcast domains exist?

    • A.Eight collision domains and eight broadcast domains.
    • B.One collision domain and eight broadcast domains.
    • C.Eight collision domains and one broadcast domain.✓ Answer
    • D.One collision domain and one broadcast domain.

    A switch buffers and forwards frames port by port, so each port is its own collision domain — that is why full duplex is possible, and why a hub, which simply repeats every signal out all ports, would leave all eight stations in one collision domain instead. Broadcast frames are flooded to every port in the VLAN, so the switch forms a single broadcast domain; dividing it requires another VLAN or a routed interface.

    Source: IEEE 802.3 collision domains and IEEE 802.1Q VLAN broadcast domainsReport a problem with this question

17. 17. A load balancer reads the requested URL path in each HTTP request and sends requests for /images to one server pool and requests for /api to another. At which OSI layer is that load balancer operating?

    • A.Layer 2, because it rewrites MAC addresses.
    • B.Layer 4, because it distributes traffic based on TCP port numbers.
    • C.Layer 3, because it chooses a destination IP address.
    • D.Layer 7, because the decision requires reading application-layer content.✓ Answer

    A URL exists only inside the HTTP payload, so the appliance must terminate the connection and parse application-layer data before it can decide, and that is Layer 7 behaviour. A Layer 4 load balancer sees only IP addresses and TCP or UDP port numbers, so it cannot distinguish one path from another on the same port. The same split explains packet-filtering firewalls operating at Layers 3 and 4 versus application-aware firewalls at Layer 7.

    Source: ISO/IEC 7498 OSI Basic Reference Model — Application layer; CompTIA Network+ Networking Concepts domainReport a problem with this question

18. 18. A security appliance is attached to a switch mirror (SPAN) port. It raises alerts about suspicious traffic but never stops it. Which appliance is this, and why can it not block?

    • A.A forward proxy, which duplicates every session it handles.
    • B.An intrusion detection system; it receives only a copy of the traffic out of band, so the original frames have already been forwarded.✓ Answer
    • C.An intrusion prevention system; it cannot block because mirror ports are read-only for the first packet of a flow.
    • D.A firewall, which is passive by definition.

    Placement determines capability. An IDS sits out of band, fed by a mirror port or a tap, and by the time it has analysed the copy the real packet is already on its way, so alerting is all it can do. An IPS is deployed inline in the traffic path, which lets it drop or reset matching traffic — and also means a false positive or an appliance failure can interrupt production traffic.

    Source: CompTIA Network+ — Networking Concepts domain (IDS and IPS placement)Report a problem with this question

19. 19. In a spine and leaf data center fabric, how are the switches interconnected and which traffic pattern is the design meant to serve?

    • A.Every leaf connects to every spine, spines do not connect to each other, and the design optimizes east-west server-to-server traffic.✓ Answer
    • B.Each leaf connects to exactly one spine in a hub-and-spoke pattern, so every path has a different hop count.
    • C.Every spine connects to every other spine and each leaf attaches to a single spine, which optimizes north-south traffic.
    • D.Leaves are cabled to their neighbors in a ring and the spine acts as a backup path for north-south traffic.

    Spine and leaf is a two-tier folded fabric: each leaf has an uplink to every spine, spines are never cabled to each other and leaves are never cabled to each other. Any server-to-server path is therefore leaf to spine to leaf, giving every pair the same predictable hop count and latency, and adding a spine adds bandwidth without redesigning the network. That equal-cost fabric exists because virtualized data centers generate far more east-west traffic between servers than north-south traffic entering and leaving the data center.

    Source: CompTIA Network+ — Networking Concepts domain (spine and leaf architecture; north-south and east-west traffic flows)Report a problem with this question

20. 20. Eight sites are to be joined in a full mesh, in which every site has a direct link to every other site. How many links are required?

    • A.28✓ Answer
    • B.56
    • C.16
    • D.8

    A full mesh requires n(n-1)/2 links, because each of the n nodes connects to the other n-1 nodes and every link is shared by two nodes: 8 x 7 / 2 = 28. The count grows with the square of the node count, which is why a full mesh is reserved for a small number of critical sites while larger networks use a partial mesh or a hub-and-spoke design.

    Source: CompTIA Network+ — Networking Concepts domain (full mesh link formula n(n-1)/2)Report a problem with this question

21. 21. A group of servers needs raw block-level storage volumes delivered over a dedicated storage network rather than file shares carried on the production LAN. Which term describes that arrangement?

    • A.A NAS, which shares files over the existing LAN using protocols such as SMB or NFS.
    • B.A PAN, because the link is a short dedicated connection.
    • C.A MAN, because the storage spans a metropolitan area.
    • D.A SAN, which presents block-level volumes to servers over a dedicated storage fabric.✓ Answer

    A storage area network presents raw block devices, so each server sees a volume that it formats with its own file system, and it runs on a separate fabric such as Fibre Channel or an isolated iSCSI network so that storage traffic does not compete with user traffic. Network attached storage is the file-level alternative: it owns the file system and serves files over the ordinary LAN. MAN and PAN describe geographic scope, not a storage model.

    Source: CompTIA Network+ — Networking Concepts domain (SAN block-level storage versus NAS file-level storage)Report a problem with this question

Practice questions based on the CompTIA Network+ exam domains and the published standards the exam is built on (IEEE 802.3 and 802.11, IEEE 802.1Q and 802.1X, the IANA port registry, IETF RFCs, and TIA-568 cabling). CompTIA and Network+ are marks of CompTIA; this site is not affiliated with or endorsed by CompTIA. Exam objectives are revised periodically — check the current official objectives for your exam version before testing. About the Network+ exam →