Wired network access

There’s one common network building block we haven’t covered in detail yet: the Local Area Network (LAN).

A LAN is a group of devices connected to each other within a relatively small area (for example, a home, office floor, or building). A LAN is usually part of a larger network design - such as a SOHO network or a campus network - rather than a complete architecture by itself.

Modern LANs typically fall into one of these two types:

• Ethernet LANs: groups of wired devices connected through an Ethernet switch (or multiple interconnected switches)
• Wireless LANs (WLANs): groups of wireless devices connected through a wireless access point (AP) (or multiple interconnected APs)

We will discuss Wireless LAN connectivity in more detail later in this book.

To access a network, endpoints usually connect first to a wired or wireless LAN. Most LANs also include a router, which allows devices inside the LAN to communicate with devices outside the LAN.

The following image shows an example of a typical Ethernet LAN.

(Image 022)

Image title - Typical LAN

As you can see, endpoints plug into a switch. The switch forwards Ethernet frames between endpoints inside the LAN. If you have many endpoints, you may need additional switches. Those switches connect to each other using the designs discussed in Chapter 1.6.1 Network Topology Architectures.

The switch then connects to a router. The router forwards (routes) packets when an endpoint’s destination is outside the LAN.

Ethernet networks are categorized as a star topology because, in a logical diagram, endpoints connect outward from a central Ethernet switch. The switch is the center of the “star,” and each endpoint connection radiates away from it.

Image

Cables for wired network access

To connect devices in a wired LAN, you’ll use network cabling. Many cabling standards exist, but they generally fall into two categories:

• Copper cables
• Fiber optic cables

This discussion focuses on Layer 1 (the Physical Layer) of the TCP/IP (or OSI) model.

Copper Ethernet cables

The most common Ethernet cables today are made of twisted pairs of copper wires inside a plastic jacket. The wire ends (often called the pinouts) fit into a clear plastic connector called an RJ-45 connector, which is crimped onto the cable.

(Image 023)

Image title - Ethernet Cable

Copper twisted-pair Ethernet cables come in two main types:

• Unshielded Twisted Pair (UTP)
• Shielded Twisted Pair

The difference is that shielded twisted pair adds aluminum foil shielding around the twisted pairs before they’re bundled inside the outer jacket. Shielding helps reduce electro-magnetic interference (EMI), which can cause data to be misinterpreted or lost while traveling over the cable. UTP is more common in modern networks.

(Image 024)

Image title - UTP vs Shielded Twisted Pair

In a typical wired LAN connection:

• One end of the cable plugs into the endpoint’s Network Interface Card (NIC Card).
• The other end plugs into a switch.

The port on a NIC that accepts an Ethernet copper cable is often called an RJ-45 port. When the NIC pins and the switchport pins contact the cable’s pins, the devices exchange electrical signals. Network devices interpret those signals as bits (1s and 0s), which represent data.

(Image 025)

Image title - Endpoints connecting to a Switch via an Ethernet Cable

Copper Ethernet has a practical distance limit. In most Ethernet copper standards, the maximum cable length is 100 meters (328 ft) before signal quality begins to degrade.

Twisted-pair copper Ethernet cables are also used to connect network devices to each other, such as:

• Switch to switch
• Switch to router
• Router to router

(Image 026)

Image title - Switches connecting to Switches and Routers connecting to Routers

When you connect an endpoint to a switch (or directly to a router), the pinout arrangement is typically the same on both ends of the cable. This is called a straight-through cable.

(Image 027)

Image title - Ethernet Straight Through Cable

When you connect like devices to each other, (e.g. switch to switch, router to router or PC to PC) certain wire pairs traditionally needed to be crossed so transmit pins on one side line up with receive pins on the other side. This is called a crossover cable.

(Image 028)

Image title - Ethernet Crossover Cable

The reason is that endpoints typically have the same transmit/receive pin arrangement on their NICs. Without crossing the appropriate pairs, both devices may transmit on the same wire pairs and listen on the same wire pairs, so the transmitted signals won’t arrive on the correct receiving pins.

(Image 029)

Image title - The need for crossover cables

This isn’t required when connecting to a switch because switchports are designed with the necessary transmit/receive pairing in mind.

It’s also worth noting that many modern devices support Auto-MDIX, which automatically detects whether the connected device uses the same pin arrangement. If needed, the device logically swaps which pins it uses for transmit and receive. With Auto-MDIX, a straight-through cable can often be used even for like-to-like connections (for example, PC to PC, switch to switch, or router to router).

Duplex

A device may or may not be able to transmit and receive at the same time. This capability is called duplex.

• Half duplex: the device can either transmit or receive at a given moment, but not both
• Full duplex: the device can transmit and receive simultaneously

Most modern network devices operate in full duplex.

When an endpoint connects to a switch or router, both sides must use the same duplex mode. A duplex mismatch can prevent proper communication. Duplex can be configured manually or negotiated automatically between devices.

Ethernet cabling standards and speeds

Ethernet cable standards determine two key limits:

• Distance: how far the electrical signal can travel over copper
• Bandwidth: how much data can be transferred per second (commonly measured in bits per second)

Here are the common copper Ethernet standards and what they support.

Fast Ethernet

Fast Ethernet provides 100 Megabits per second of bandwidth over UTP.

• 100 Megabits per second is abbreviated as 100Mbps.
• The twisted-pair copper standard is 100BASE-T (also referred to as 100BASE-TX).
• The common cable type for Fast Ethernet is Category 5 (Cat 5).
• Cat 5 uses the RJ-45 connector.

Despite the name, Fast Ethernet is considered obsolete in modern networks and has largely been replaced by faster standards.

Gigabit Ethernet

Gigabit Ethernet provides 1,000 Megabits per second (1000 Mbps), which equals 1 Gigabit per second (1Gbps).

• The UTP standard is 1000BASE-T (or 1000BASE-TX).
• Common cable types that support Gigabit Ethernet are Cat 5e and Cat 6.

MultiGigabit Ethernet

Beyond 1Gbps, Ethernet standards include higher speeds such as 10Gigabit Ethernet (10GBASE-T), as well as 25Gbps, 40Gbps, 100Gbps, and beyond.

As speeds increase past 1Gbps, fiber optic cabling becomes more common because it supports higher bandwidth over longer distances. Some copper standards can still support higher speeds, though.

• Copper cabling that can support 10Gbps up to 40Gbps includes Cat 6a, Cat 7, and Cat 8 (respectively).

The following is a table explaining the various copper Ethernet standards, speeds and their respective distance limitations (Cat 7 and Cat 8 are omitted as they are out of scope of the CCNA exam):

Copper Ethernet table

Category	Ethernet standard	Bandwidth	Distance
Cat 5	100BASE-T, 100BASE-TX	100Mbps	100m (328 ft)
Cat 5e	1000BASE-T, 1000BASE-TX	100Mbps, 1Gbps	100m (328 ft)
Cat 6	1000BASE-T	1Gbps	100m (328 ft)
Cat 6a	1000BASE-T, 10GBASE-T	1 Gbps, 10Gbps	100m (328 ft)

Fiber Optic Ethernet cable standards and speeds

Fiber optic cables have been used for a long time and are commonly used to interconnect routers across the backbone of the internet.

Because copper (UTP) has distance limitations, fiber optic cabling is typically used when you need to connect devices over longer distances. That’s why fiber is common in WAN (Wide Area Network) connections.

Fiber is also widely used for device-to-device connections (for example, switch to switch and router to router) because:

• It supports higher bandwidth.
• It isn’t susceptible to electro-magnetic interference (EMI) the way copper is.
• It carries data as pulses of light rather than electrical signals.

A fiber optic cable contains very thin strands of glass that guide light. The main parts are:

• Core: the center of the strand where light travels
• Cladding: a surrounding glass layer that reflects light back into the core
• Coating/Buffer: a protective layer around the glass
• Strength member: reinforcement (often Kevlar or fiberglass) to reduce damage from pulling, bending, or pinching
• Jacket: the outer PVC layer

Even with reinforcement, fiber cables are still relatively fragile and can be damaged if handled roughly.

(Image 030)

Image title - Fiber Optic Cable

Fiber optic connections typically use two strands:

• One strand transmits data.
• The other strand receives data.

When you connect fiber to a device, you need to make sure the transmit side on one end connects to the receive side on the other end.

Fiber uses several connector types, but the one you’ll most commonly see as a CCNA candidate is the LC (Lucent) connector.

(Image 031)

Image title - LC Connector

Fiber optic cables don’t plug directly into most network devices. Instead, you plug the fiber into a transceiver, also called an SFP (Small Form-factor Pluggable). The transceiver converts light pulses into electrical signals that the device can process.

SFPs plug into dedicated SFP ports on network devices. They are hot-pluggable, meaning you can insert or remove them without restarting the device.

(Image 032)

Image title - SFP (Transceiver)

There are several fiber standards, but two of the most common are below.

1000BASE-SX

1000BASE-SX runs at 1Gbps for distances up to 550 meters.

• It uses multimode fiber.
• Older multimode fiber (OM1/OM2) often has an orange jacket.
• Newer multimode fiber (OM3/OM4) often has an aqua/teal jacket.

Newer multimode fiber types can also support higher speeds.

1000BASE-LX

1000BASE-LX runs at 1Gbps for distances up to 5 kilometers (3 miles).

• It’s commonly used for WAN links and long-distance connections.
• It uses single mode fiber.
• Single mode fiber typically has a yellow jacket.

10GBASE-SR and 10GBASE-LR

These standards support up to 10Gbps:

• 10GBASE-SR: up to 400 meters using multimode fiber
• 10GBASE-LR: up to 6 miles using single mode fiber

Fiber Optic Cables can obviously go up to 100Gbps, 400Gbps or even up to 1.2 Terabits per second however, those higher speeds are out of the scope of the CCNA exam.

The following is a table explaining the various Fiber Optic Ethernet standards, speeds and their respective distance limitations:

Fiber Optic Ethernet table

Optical media	Ethernet standard	Bandwidth	Distance
Multimode	1000BASE-SX	1Gbps	550m (1804 ft)
Single Mode	1000BASE-LX	1Gbps	5km (3 miles)
Multimode	10GBASE-SR	10Gbps	400m (1312 ft)
Single Mode	10GBASE-LR	10Gbps	10km (6 miles)