Computer Networks: Architectures, Topologies, and Components

Peer-to-Peer Architecture

• Peer-to-peer (P2P) networks operate on a decentralized model where every participating workstation (peer) possesses identical status, authority, and system privileges.
• Functional comparison with Client-Server architecture:
  • In client-server networks, distinct roles are maintained: a dedicated server provisions network services and resources, whereas clients initiate requests to consume those services.
  • In peer-to-peer networks, no hardware role differentiation exists; any device on the network can concurrently function as a service provider (server) and a service requester (client).
• Advantages of Peer-to-Peer networks:
  • Simple Configuration: Device configuration and network setup are straightforward, requiring only direct physical or logical interconnection without complex centralized coordination.
  • Low Cost: P2P networks are significantly less expensive than client-server setups because they eliminate the requirement for costly, dedicated centralized servers and specialized network management infrastructure.
• Disadvantages of Peer-to-Peer networks:
  • Poor Scalability: As the network increases in node count, centralized administration and management become increasingly difficult, leading to a loss of network control.
  • Inadequate Security and Lack of Central Management:
    • Security cannot be universally enforced due to the complete absence of centralized administrative points.
    • Decentralized Entry Points: In a P2P network containing $5$ workstations, each individual workstation may establish its own autonomous connection to the Internet. Consequently, administrators must secure $5$ separate perimeter entry points rather than a single, controlled gateway point.
    • Unrestricted Peer Access: When a workstation configures a resource (such as a shared folder under a specific name), any peer device across the network can access that workstation directly. This exposure creates severe security vulnerabilities compared to client-server systems, where clients are isolated from one another and can only access the server.

Network Classification by Geographical Size and Area

• Networks are divided into three major categories based strictly on geographic coverage area:
  • Local Area Network (LAN):
    • The smallest foundational category of network infrastructure.
    • Geographical Span: Covers a single, limited geographic area, ranging in size from a single room or laboratory to an individual office building or an entire university campus.
    • Administrative Control: Owned, operated, and maintained by a single unified organization, enterprise, or institution.
    • Operational Performance: Engineered to deliver high data transmission speeds and localized service delivery for institutional applications.
  • Metropolitan Area Network (MAN):
    • Intermediate classification: Larger than a LAN and smaller than a WAN.
    • Geographical Span: Spans the physical boundaries of an entire municipality, city, or a large enterprise maintaining interconnected branch offices across an urban area.
    • Structural Composition: Constructed by interconnecting multiple LANs together. Aggregating $5$ to $10$ LANs forms a MAN, and uniting up to $20$ distinct LANs establishes a regional metropolitan network.
  • Wide Area Network (WAN):
    • The largest classification of network infrastructure.
    • Geographical Span: Extends across extensive physical boundaries, covering states, nations, continents, or the entire globe.
    • Structural Composition: Composed entirely of interconnected LANs aggregated on a massive scale. Grouping $100$, $200$, or $300$ distinct LANs together establishes a WAN.
• Foundational Role of the LAN:
  • The fundamental building block of all computer networking is the Local Area Network.
  • Higher-tier networks (MANs and WANs) do not differ from LANs in fundamental hardware mechanics or internal server operations; they are vast aggregations of interconnected LAN infrastructures.

Network Classification by Physical Topology

• Network topology designates the physical layout, structural visual arrangement, and interconnected layout of physical network nodes.
• Bus Topology:
  • Physical Structure: Devices are connected sequentially along a single continuous linear transmission path.
  • Cabling and Hardware: Operates over a single shared physical cable connecting all network nodes directly, without intermediary connectivity devices such as hubs or switches.
  • Shared Medium: All connected stations share a single communication channel.
  • Advantages: Inexpensive to purchase, install, and implement.
  • Disadvantages:
    • Limited Scalability: Expanding the bus by attaching additional nodes weakens signal strength due to physical attenuation.
    • Difficult Troubleshooting: Diagnosing, isolating, and fixing cable fractures or physical signal interruptions is difficult and time-intensive.
    • Lack of Fault Tolerance: The topology is not fault-tolerant; a break or electrical failure anywhere along the central cable halts communication across the entire network.
• Ring Topology:
  • Physical Structure: Every device is wired directly to two neighboring devices, forming an unbroken closed circular loop.
  • Data Directionality: Transmissions travel unidirectionally around the ring in a clockwise data transmission direction.
  • Disadvantages:
    • Node Failure Vulnerability: A hardware malfunction or physical disconnection of a single node can disable the entire network.
    • Inflexible Modification: Inserting new nodes or reorganizing the physical network is difficult and requires disruption to the loop.
• Star Topology:
  • Physical Structure: Every individual node connects independently to a central intermediary networking device (such as a router or switch).
  • Physical Media: Utilizes point-to-point dedicated cable runs per node, typically consisting of twisted-pair copper wire or fiber optic cabling.
  • Advantages:
    • High Fault Tolerance: The failure, disconnection, or damage of an individual node or cable run does not interrupt network traffic among the remaining active nodes.
    • High Flexibility: Workstations can be attached or removed seamlessly without bringing down the network.
    • Industry Benchmark: Represents the premier and most universally deployed physical network topology in modern enterprise environments.
• Hybrid Topology:
  • Integrates two or more distinct physical network topologies into a single heterogeneous network layout.

Modern Industry Standards and Practice

• Professional networking and cybersecurity practices focus predominantly on three interconnected standards:
  • By Role: The Client-Server architecture represents the universal global computing model. Global web services (including platforms like Facebook, Google, and Yahoo) operate entirely through client-server paradigms.
  • By Geographic Scope: The Local Area Network (LAN) serves as the primary foundational administrative domain.
  • By Physical Structure: The Star Topology is the dominant physical layout deployed in commercial and enterprise environments.

Core Components of Client-Server Networks

• A functional client-server network infrastructure requires four fundamental components:
• 1. The Client:
  • Hardware devices operated by end-users (e.g., desktop computers, laptops, smartphones).
  • Function: Submits requests across the network medium to obtain files, data, computational processes, or centralized services from a server.
• 2. The Server:
  • Function: Manages and arbitrates centralized resource sharing, including print services, file storage, operational applications, and centralized database systems.
  • Hardware Characteristics: Contains substantial processing power (multi-core high-throughput CPUs), large memory capacity (RAM), and high-capacity storage arrays (hard drives) to support concurrent network loads.
  • Cost Differential: High-capacity commercial servers can cost $20{,}000\,\text{dinars}$, whereas an individual client endpoint may cost $100\,\text{dinars}$.
  • Sizing and Form Factors:
    • Desktop Tower Chassis: Sized similarly to a standard desktop PC case, providing server capabilities sufficient to support a small business of approximately $20$ employees.
    • Dual/Multi-Chassis Workstations: Expanded desktop-style chassis setups supporting medium workgroups of up to $200$ employees.
    • Server Rooms: Specialized rooms containing server racks built to support large organizations, enterprises, or government ministries servicing $8{,}000$ to $100{,}000$ users.
    • Industrial Data Centers: Expansive, warehouse-scale facilities located in cold climates (such as Northern Canada, Northern Europe, and Siberia) to assist in thermal cooling. These facilities are large enough to drive vehicles inside and contain warehouse floors of servers engineered to process data for billions of people.
• 3. Network Interface Card (NIC):
  • Hardware device installed inside the client or server motherboard to provide the physical connection to the transmission medium.
  • Supports transmission interfaces via physical Ethernet copper cabling (RJ45), wireless radio frequency transceivers (Wi-Fi), or hybrid configurations incorporating both wired and wireless communication interfaces.
  • Function: Acts as the physical and data-link interface that translates internal system data into network-compatible physical signals, enabling communication between hosts across the network.
• 4. Network Operating System (NOS):
  • Standard operating systems (such as desktop Windows) are designed for local user applications and client-side processing, lacking the software infrastructure required to arbitrate server services.
  • Function: Specialized operating systems built to act as the intermediary interface between network hardware (servers, switches, routers) and network clients, directing multi-user traffic, security rules, and centralized access.
  • Examples: Windows Server 2003, UNIX-based server environments, and Linux server operating systems.

Questions and Discussion

• Comparison Between Star and Hybrid Topologies:
  • Question: Is a Star topology identical to a Hybrid topology, or are they different?
  • Response: They are distinct. A Star topology connects all devices individually to a single central device, whereas a Hybrid topology combines two or more different topologies (such as integrating Star and Bus layouts) within a single network installation.
• Review of the Three Primary Network Categories:
  • Prompt: What are the three primary criteria used to classify computer networks?
  • Response: Mus'ab identified the three classifications:
    • First: By the role of the device within the network (Client-Server versus Peer-to-Peer).
    • Second: By geographical coverage area (Local Area Network [LAN], Metropolitan Area Network [MAN], and Wide Area Network [WAN]).
    • Third: By the physical shape and layout of the network nodes (Physical Topology: Bus, Ring, and Star).
• Server Dimensions and Scale Variations:
  • Prompt: What are the physical dimensions and structural configurations of network servers?
  • Response: Awad noted that an individual server installation can occupy an entire room. Further clarification established that physical server scale depends entirely on client capacity: serving $20$ users requires a single desktop-sized server case; $200$ users requires paired hardware chassis; $8{,}000$ to $100{,}000$ users requires full server rooms; and global operations servicing millions to billions of users require warehouse-scale data center facilities.