Week 2 · Lesson 7

Reliable Networks

A network that works when everything is fine is not enough. Real networks must keep working when a link snaps, when the business grows, when voice and video compete with file transfers, and when someone tries to break in. These needs are captured by four characteristics every reliable network must have.

CCNA 200-301 Domain 1.0 Network Fundamentals Verified 2026-10-11
Part 1

The four characteristics

Cisco describes a reliable network with four characteristics. Learn them as a set; exam questions often give a scenario and ask which characteristic it demonstrates.

The four characteristics · select one

In one line. Fault tolerance keeps the network up, scalability keeps it growing, QoS keeps it smooth, and security keeps it safe.
CharacteristicQuestion it answersCore tool
Fault toleranceWhat happens when something breaks?Redundancy and alternate paths
ScalabilityWhat happens when we grow?Modular, hierarchical, standards-based design
Quality of serviceWhat happens when traffic competes?Classify and prioritise delay-sensitive traffic
SecurityWhat happens when someone attacks?Confidentiality, integrity, availability plus access control
Part 2

Fault tolerance

Fault tolerance is the ability of a network to keep working, or to recover quickly, when a component fails. The design principle behind it is simple but powerful: provide redundancy in hardware and in paths, so that no single failure can bring the network down. Limiting the number of devices affected by a failure is the goal — not preventing every failure.

Single point of failure

Any one component whose failure disconnects the network. Fault-tolerant design removes these.

Redundancy

Duplicate components — a second switch, router, link, or power supply — that can take over.

Alternate paths

More than one route between important points, so a broken link can be bypassed.

Fault tolerance rests on packet switching. Data is divided into packets, and each packet is forwarded independently toward the destination. If one path fails, packets can be sent along an alternate path instead. This is why packet-switched networks are inherently more resilient than a system that ties up one dedicated circuit for the whole conversation.

Rule of thumb. Ask “what breaks if this one thing fails?” If the answer is “everything,” you have found a single point of failure and need redundancy there.
Part 3

Redundant designs

A redundant design deliberately duplicates the parts of the network that matter most. Compare a single-link chain with a redundant mesh, and then break each one to see the difference.

Single link vs redundant design · step through the failure
1 / 4
Single link PC R1 Server one failure = outage single point of failure Redundant design PC R1 R2 Server alternate path keeps service up

Part 4

Scalability

Scalability is the ability of a network to grow — more users, more devices, more sites, more traffic, and new services — without a drop in performance and without having to be redesigned from scratch. A scalable network is built the way a city grows: with a clear structure that new pieces attach to.

The test of scalability is that yesterday’s design still works tomorrow when the network is several times larger. In a flat design every device would need a link to every other device, so the link count — and the cost, and the configuration work — would grow far faster than the network itself. A hierarchy keeps the growth proportional.

Part 5

Quality of service

Modern networks are converged: they carry data, voice, and video together. But these traffic types tolerate delay and loss very differently. Voice and video are real-time and delay-sensitive; email and file transfers are not. Without management, a large backup could consume all the bandwidth and wreck a simultaneous voice call.

Definition — quality of service (QoS). The set of techniques that let a network treat different traffic differently, giving priority to delay-sensitive traffic so it is delivered in time even when the network is busy.

The essential QoS steps are to classify traffic into categories (for example voice, video, best-effort data) and mark it, then to apply a policy such as giving the real-time classes priority queues and guaranteed minimum bandwidth. The result is smooth voice and video despite congestion.

Load on the link · see what QoS protects

Part 6

Security as a design requirement

Security means protecting the network, the data on it, and the people who use it. The goals are usually expressed as the CIA triad.

GoalMeaningExample control
ConfidentialityOnly the intended recipients can read the data.Encryption
IntegrityData is not altered in transit or storage.Checksums, digital signatures
AvailabilityThe network and its services are usable when authorised users need them.Redundancy, backups, DDoS defences

Access to resources is controlled by the authentication, authorization, and accounting (AAA) framework:

These goals are implemented with tools such as firewalls, intrusion prevention systems, encryption, access control lists, and secure device access. Security is a later, deep topic; for now, the point is that it is a design requirement of a reliable network, not an afterthought.

Part 7

Recovering from a failure

Once a redundant path exists, the network must actually use it. When a link or device fails, the routers and switches that know about both paths must detect the change and agree on the new best route. This process is called convergence. During convergence some traffic may be briefly disrupted while devices update their forwarding information. Fast convergence — converging in a fraction of a second rather than many seconds — is what keeps a failover almost invisible to users.

So fault tolerance is really two things working together: redundancy gives the network a spare path, and fast convergence lets the network switch to it quickly.

Exam tip — which pillar is it? Read the scenario for its keyword: a second link, an alternate route, no single point of failure = fault tolerance; adding more users, sites, or devices without redesign, modular growth = scalability; voice or video priority, classifying and marking traffic = QoS; passwords, encryption, keeping unauthorised users out = security. If the question mentions the network “continuing to operate” after a failure, the answer is fault tolerance.
Which pillar is it? · read the scenario

ScenarioCharacteristic
A switch has a duplicate power supply.Fault tolerance
Two ISPs are used so one provider outage does not disconnect the office.Fault tolerance
A company adds a new floor of users without changing the core design.Scalability
Equipment from several vendors interoperates using open standards.Scalability
Voice packets are placed in a priority queue ahead of file downloads.Quality of service
Traffic is classified as voice, video, or best-effort and marked accordingly.Quality of service
A firewall blocks unauthorised inbound connections.Security
A user is asked to log in and is only allowed the resources their role permits.Security
Part 8 · Your turn

Review

Check what you have learned with these review questions.

Review · 6 questions

Sources

Sources & references

  1. Cisco. CCNA (200-301) Exam Topics, Network Fundamentals (reliability and network architectures). cisco.com.
  2. Cisco Networking Academy. CCNA: Introduction to Networks, module “Networking Today” (network reliability characteristics). netacad.com.
  3. Nichols, K. & Jacobson, V. RFC 2474: Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers (1998). rfc-editor.org/rfc/rfc2474.
  4. Meyer, D. RFC 2475: An Architecture for Differentiated Services (1998). rfc-editor.org/rfc/rfc2475.
  5. Stallings, W. Network Security Essentials (confidentiality, integrity, availability). cisco.com.

Target exams: CCNA 200-301 v1.1 · CyberOps Associate 200-201. Last content verification: 2026-10-11.