Understanding Network Components

• Network connections extend to both wireless and wired networks.

Network Components

• Clients are devices used to access the network, these include workstations, laptops, and tablets.
• Servers provide resources to the network, such as email servers and file servers.
• Hubs connect devices but aren't commonly used because of their limitations.
• Switches are smarter hubs that ensure security and efficient bandwidth utilization.
• Wireless Access Points (WAPs) enable wireless devices to connect to a wired network using radio frequency waves.
• Routers connect different networks and make intelligent forwarding decisions based on IP addresses.
• Firewalls act as a security barrier between an internal network and the internet, monitoring and controlling traffic.
• Load Balancers distribute network/application traffic across servers, preventing bottlenecks.
• Proxy Servers act as intermediaries between user devices and the internet, which enhances security and privacy.
• Intrusion Detection Systems (IDS) detect unauthorized accesses or anomalies.
• Intrusion Prevention Systems (IPS) detect and take action to prevent intrusions.
• Controllers manage flow control in software-defined networking (SDN), offering flexibility and efficiency.
• Network-attached Storage (NAS) Devices provide dedicated file storage systems and data access to authorized clients.
• Storage Area Networks (SANs) have high-speed networks for consolidated block-level data storage, which enhances accessibility.
• Media includes physical materials for data transmission, such as copper cables and fibre optic cables.
• Wide Area Network (WAN) Links connect networks over large areas and are essential for global connectivity.
• Understanding network components are crucial for efficient and secure data transmission in information technology, which aids in network design, management, problem-solving, and security implementation.

Network Resources

• The Client/Server Model utilizes a dedicated server for centralized access to files, scanners, printers, and resources.
• Centralized administration, easier management, and better scalability are benefits of the Client/Server Model.
• Higher cost and the requirement for dedicated hardware and a specialized skillset are drawbacks of the Client/Server Model.
• The Client/Server Model is a leading model in business networks.
• The Peer-to-Peer Model provides direct sharing of resources among peers (laptops, desktops).
• Difficult administration and backup due to dispersed files are drawbacks of the Peer-to-Peer Model.
• Redundancy, complex management, and scalability issues are further drawbacks of the Peer-to-Peer Model.
• Low cost and no specialized infrastructure or hardware are benefits of the Peer-to-Peer Model.
• Decentralized management and poor scalability for large networks are drawbacks of the Peer-to-Peer Model.
• The Peer-to-Peer Model is not recommended for business networks.

Network Geography

• A Personal Area Network (PAN) is the smallest network type, this covers about 10 feet or less with examples such as Bluetooth and USB.
• A Local Area Network (LAN) is common in office buildings with a limited distance of up to 100 meters.
• WiFi (IEEE 802.11) or Ethernet (IEEE 802.3) can be used in LANs, with examples including office, school, and home networks.
• A Campus Area Network (CAN) is a building-centric LAN that spans numerous buildings in an area, covering several miles, with examples such as college campuses, business parks, and military bases.
• A Metropolitan Area Network (MAN) connects locations across the entire city, is larger than a CAN and is up to 25 miles.
• Examples of MANs are city departments and multiple campuses in a city.
• A Wide Area Network (WAN) connects geographically disparate internal networks, providing large geographic coverage.
• WANs can consist of lease lines or VPNs and examples are Internet, private connections between offices across countries or globally.
• PAN uses Bluetooth/USB, LAN uses IEEE 802.3 (Ethernet), CAN connects multiple LANs, MAN spans an entire city, and WAN encompasses large geographic areas.

Wired Network Topology

• Network topology is the arrangement of elements in a computer network; it includes links, nodes, clients, and servers.
• Physical topology describes physical cabling and device connections and represents real-world layouts using floorplans.
• Logical topology describes how data flows in the network and focuses on the logical connection rather than the physical placement.

Six Wired Network Topologies

• Point-to-Point Topology has a direct connection between two devices, it is simple and reliable for small-scale connections but is not scalable.
• Ring Topology has a circular data path, with unidirectional flow to prevent collisions.
• Ring Topology creates a single point of failure situation unless there are redundant connections for failover; frequently used in FDDI for long-distance fibre optic networks.
• Bus Topology connects all devices to a central cable (bus), data is accessible to all but only the intended recipient processes it; easy to install, but the entire network fails if the main cable fails.
• Collision likelihood increases with device count on a bus topology.
• Star Topology connects each node to a central point (network switch), it is robust, but the network depends on the central point's functionality.
• Hub-and-Spoke Topology is a star topology variation with a central hub connected to multiple spokes; cheaper for large networks.
• Mesh Topology is a point-to-point connection between every device for redundancy, providing robustness but it can be costly and complex. -A full mesh connects every node to every other, and a partial mesh only connects some.

Wireless Network Topology

• Infrastructure Mode utilizes a centralized wireless network with a wireless access point, it is common in homes and similar to a star topology in a physical network.
• Ad Hoc Mode is a decentralized wireless network that operates like a peer-to-peer network, having dynamic routing decisions made on the fly, and allows for dynamic joining and leaving of devices.
• Devices connect directly using Ad Hoc Mode, resembling old-school chat rooms
• Wireless Mesh has unique interconnections of different nodes, devices, and radios, creating a mesh topology for expansion and redundancy.
• Bluetooth, WiFi, microwave, cellular, and satellite can be combined with Wireless Mesh.
• Enables large-scale network access in harsh environments.
• Uses different radio frequencies to establish reliable connections.
• Wireless Mesh is used for post-disaster scenarios and humanitarian assistance missions.

Datacenter Topology

• A Datacenter is composed of networked computers and storage to organize, process, store, and disseminate large amounts of data.
• The three-tiered hierarchy is comprised of core, distribution/aggregation, and access/edge layers.
• The Core Layer houses high-performance routers and merges geographically separated networks.
• The Distribution/Aggregation Layer provides boundary definition and defines policies for the network through access lists and filters.
• The Access/Edge Layer connects endpoint devices using regular switches.
• Having 3 layers leads to better performance, management, scalabililty, redundancy and helps with troubleshooting.
• The Collapsed Core network architecture merges the core and distribution layers into a single layer.
• Creates a two tiered core and is a simplified architecture for medium to small datacenters/
• Spine and Leaf Architecture is an alternative type of network architecture that focuses on communication within datacenters, it consists of leaf and swine switching layers.
• The Leaf Layer consists of access switches that aggregate traffic from servers and then connect directly into the spine layer.
• The Spine Layer's switches interconnect all leaf layer switches into a full-mesh topology.
• Enhances speed and reduces latency compared to a three-tiered hierarchy.
• Works well with a Software Defined Networks (SDN) and can be used along with standard three-tiered hierarchies.
• North-South traffic enters or leaves data center from outside systems, while East-West traffic refers to data flow within a datacenter.

OSI Model

• The Open Systems Interconnect Model (OSI) model was developed in 1977 and is a reference model used to categorize the functions of a network and help with troubleshooting.
• Networks today operate under the TCP/IP mode.
• Networks are designed to get data across networks.
• Layers in order are: Physical, Data Link, Network, Transport, Session, Presentation, and Application.
• Data stream names in order are: Bits, Frames, Packets, Segments, Data, Data, Data.

Layer 1

• Layer 1 (Physical) is responsible for the transmission of bits across the network, including physical and electrical network characteristics.
• Data type here is bits, represented as a series of 1s and 0s.
• Transition modulation switches between levels to represent 1 or 0; copper wire uses voltage and fiber optic cable uses light.
• RJ-45 Connector is used in CAT5/CAT6 cables, also includes wiring standards and pin arrangements (TIA/EIA-568A/B).
• Physical topology arranges physical network layouts connected using cable.
• There is asynchronous communication (start and stop bits for out-of-sync data transmission) and synchronous communication (real-time communication with a common time source).
• Broadband divides bandwidth into separate channels while baseband uses the entire cable frequency.

Multiplexing

• Multiplexing involves taking some limited amount of resource and using if more efficiently, and allows multiple people to use a baseband connection.
• Time Division (TDM) allocates dedicated time slots, StatTDM dynamically allocates based on need, and Frequency Division (FDM) divides the channel.
• Cables, Ethernet, and Coaxial are media, while Bluetooth, Wi-Fi, and Near Field are Wireless.
• Hubs, Access points and media converters are infrastructure devices for network connections on layer 1.
• Device characteristics for Layer 1 simply repeat/regenerate whatever they receive with no logic.

L2 Data Link Model

• L2 packs layer 1 bits into frames and then transmit it across a network. It performs error correction, identifies MAC address for flow control.
• MAC Address is 48 bits assigned to every network interface card, or NIC, made and is written in hexadecimal.
• An MAC address is used for physical identification allowing a device to operate with a logical address.
• The first 24 bits identify the manufacturer, and the remaining 24 bits specify the device.
• Logical Link Control (LLC) provides connection services, assures message receipts, and controls data flow.
• Isochronous Mode uses a common reference clock with less overhead, while synchronous methods utilize the same clock and asynchronous devices reference clock cycles.
• NICs and Bridges operate on Layer 2.

Layer 3

• Layer 3 (Network) forwards traffic through a network using a logical address.
• IP comes in 2 Variants, IPV4 and IPV6. IPV4 is written in dotted octet notation.
• Common protocols for Layer 3: AppleTalk and IPX.
• Packets are split at this layer switching/routing.
• Dynamic protocols like RIP and OSPF enable routers to share and update network information on best route. Connection services, augmentation and flow control are included here.
• ICMP or Internet Control Message Protocol sends error messages to an IP address from routing.
• Commonly used for the PING command, and traceroute that reads the route for a packet across the network .
• Routers use a multi layer switch at layer 3
• IPv4, IPv6 and ICMP protocols are used.
• IP and routers are commonly encountered.

Layer 4

• Layer 4 (Transport) divides the upper and lower layers of the OSI model, where the applications use transport, presentation, session.
• Data Type is Segments.
• Protocols are TCP or Transmission Control Protocol, a connection oriented that provides reliable transfer through a 3 way handshake and acknowledgement: SYN/SYN-ACK/ACK. This uses windowing for flow control across networks.
• UDP is connectionless, no handshake, for audio and visual streaming.
• Segment for TCP, Datagram for UDP.
• TCP: Reliable, three way handshake, connection oriented, segmentation, sequncing, acknowledgements.
• UDP: Connectionless- no 3 way handshake, no retransmision is needed.
• Layer 4 devices include: TCP/UDP, Wan accelerators, Load balancers and firewalls.

Layer 5

• Layer 5 (Session) manages sessions, ensuring different applications are not intertwined.
• Setting up a session checks user credentials, breaks up with bad signal, data acknowledgment needs acknowledgements to establish connection.
• Setting up, maintaining and disconnecting connections are all functions of Layer 5.
• H.323 and NetBIOS protocols, are used for voice and video and file sharing respectively.

Layer 6

• Layer 6 (Presentation) is responsible for formatting data, security, and encryption of data and communications.
• Formatting maintains compatibility of data being transmitted using the American Standard Code, or ASCII.
• Data and syntax of layer 7 gets negotiated here for formatting.
• Scripting languages control how the data looks.
• Encryption is done by Transport Layer Security is done here.
• Movie file formats are formatted in 1s and 0s in a watchable video.
• SSL secures connection, TLS maintains encrypted tunnel.

Layer 7

• Layer 7 (Application) provides a low level service where users communicate with computers.
• Application services unites components in a network app.
• Application services has 3 protocols: file transfer, file sharing, and email.
• Low level protocols: POPS, IMAP, SMTP.
• Services send announcements on devices here.
• Printers use an active directory here.
• HTTPS, and DNS web browsing domains are used.
• Remote access is implemented here using: Telnet, SSH, and SNMP.
• Encapsulation appends and removes details of header traffic is done here.
• Moving Down 7-1 does encapsulation, up the layers does decapsulation.

OSI Data Units

• Protocol Data Units (PDUs) are single units of info transmitted. Special info for 1,2,3,4 layers. Where 7 is written out in a PDU, layers 1-4 has respective names.

Headers

• TCP Header is 20 bytes, it has source-destinations ports, sequence number, and control flag: SYN-ACK, TCP Check, Window Sizes and Urgent Pointers.
• UDP Headers is 8 bytes, it has source-destinations ports, checksum, packet sizes.
• IP Headers also have various field, and fragmented off set.
• Ethernet Header has MAC addresses and ether type fields.

A Frame

• A frame being sent at Layer 2 will also contain a payload
• Can be data being sent across network, with variable byts across the networks.
• Jumbo frame: 1500 kb
• All layer has respective source/ destinations, ports, and addresses related to there transmission.
• Encapsulation happens until final has is reached.
• Dcapsulate Layer 7 until its in the right format.