Assignment -1

NAME : T. VIJAY KUMAR

Admission No :22001A0560
Class : B.Tech III year, I-sem
Branch : Computer Science And Engineering
Subject : Compter Networks
Topic : Design Of Internet
NETWORK MODELS IN DESINGING FOR AN INTERNET

Introduction

"Designing an Internet," written by David D. Clark, delves into various internet architectures, highlighting
their development, philosophy, and innovations. This chapter provides an overview of significant internet
architectures, including the DARPA Internet Protocols, alternative network architectures like Plutarch and
Triad, and concepts from the NSF Future Internet Architecture Projects. It also explores the foundational
elements and the evolution of ARPANET, along with security and performance management in network
architecture

The Design Philosophy of the DARPA Internet Protocols (1988):

• Key Points: Generality, adaptability, security, economic viability.

• Insights: Evolution and decision-making in Internet architecture.
• Abstract: This paper outlines the design philosophy and goals of the Internet architecture, including
survivability, support for multiple communication services, distributed resource management, cost-
effectiveness, and accountability. It emphasizes packet switching, gateways, and minimal assumptions
about network functions.

DARPA Internet Protocols: DARPA Internet Protocols: TCP/IP

Goals:

• Survivability: Ensuring the network can operate under any conditions, including partial failures.
• Multiple Communication Services: Supporting a variety of services, from low-latency
communications to high-throughput data transfers.
• Network Diversity: Allowing different types of networks to interconnect and communicate.

Emphasis: The architecture focused on packet switching, using gateways to connect different networks and
minimizing assumptions about the underlying network functions to promote flexibility and adaptability.

Alternative Network Architectures:

• Triad: The Triad project explored regional diversity in architecture by supporting different network
architectures within various regions. It proposed a three-tiered network architecture to enable regional
autonomy in design and operation.(Three-tiered architecture, regional autonomy, interoperability.)
 • Plutarch: This study proposed a hierarchical architecture that allows for regional diversity in network
design while ensuring interoperability and end-to-end delivery. It aimed to provide a platform for
innovation and experimentation in network design and services.(Hierarchical architecture, regional
diversity, interoperability.)

• NSF Future Internet Architecture Projects: The NSF FIA program funded projects that developed
new network architectures to address challenges like security, scalability, and support for new
applications. Projects like Data-Oriented Network Architecture (DONA), Framework for Internet
Innovation (FII), and Internet Indirection Infrastructure (i3) proposed innovative approaches to
network design.
• Performance Management in Network Architectures: Different network architectures like ALF and
MobilityFirst emphasize improving host processing performance and optimizing the design for
performance. Management interfaces may be required to adjust parameters based on network
conditions and user requirements to ensure optimal network performance.

• Region-based Network Architectures: The concept of regions in network architecture captures

consistent control, state, or knowledge within an area. Different types of regions exist, including those
based on trust, proximity, payment, and administration. Various architectures like Plutarch, FII, and
Triad address the requirements for region-based networks and emphasize routing, addressing, and
quality of service.

• Key Requirements: Security, scalability, support for new applications.

• Projects: DONA, FII, i3, DOA.
• Focus: Data-centric networking, innovation, indirection infrastructure.
• Vision: Secure, scalable, flexible architectures for future Internet.
• Regions in Network Architecture:
• Concept: Consistent control, state, knowledge within an area.
• Types: Shared trust, physical proximity, administrative regions.
• Role: Routing, addressing, quality of service in network design.

Security and Privacy in Network Architecture:

• Threats: Malicious attacks, compromised hosts, security breaches.

• Mitigation: Strong security measures, network segmentation, access controls.Considerations:
Optimizing performance, dynamic handover, quality of service.
 • Aspects: Fault management, configuration management, accounting, security management.
• References:
• Authors: Clark, Crowcroft, Koponen, Stoica, Walfish.
• Publications: IEEE Security & Privacy, ACM SIGCOMM Computer Communication Review, CRC
Press.
• Conclusion:
• Evolution: Government funding to commercial enterprise.
• Factors: Deregulation, open standards, technological advancements.
• Continued Impact: Economics, management, security in Internet architecture.
• Further Study:
• Explore addressing, forwarding mechanisms evolution in Internet architecture.
• Understand network management, control, and performance optimization.
• Investigate alternative network architectures for future Internet needs.

ARPANET(Advanced Research Projects Agency Network):

ARPANET (Advanced Research Projects Agency Network) was the pioneering network that laid the
foundation for the modern Internet. It was developed by the United States Department of Defense's Advanced
Research Projects Agency (ARPA), later known as DARPA.

• Initiation: ARPANET was initiated in the late 1960s to explore networking technologies and to allow
researchers across different locations to share resources and communicate more effectively.

• Objective: The primary goal was to create a resilient, decentralized network that could withstand
potential disruptions, ensuring reliable communication in case of military or other emergencies.

• Packet Switching:

Packet switching was the key concept that enabled ARPANET’s operation. Here’s how it worked:

• The data being transmitted between two computers was broken down into smaller units called packets.

• Each packet contained a portion of the data with information about its destination.

• The packets were transmitted individually across the network and routed independently based on the
addressing information.

The packet-switching approach helped organizations improve their use of computing resources with multiple
packets sent and received simultaneously.
TCP/IP Protocol

TCP, which replaced the earlier NCP protocol, was vital in enabling more reliable communication over
ARPANET. The main functions of TCP/IP were:

• Providing reliable data transmission between multiple computers connected to the network.

• Introducing flow control mechanisms that prevented data overflow and network congestion.

• Checking for errors to ensure the integrity of the received data and prevent the transmission of corrupt
data packets.

• Providing more flexibility than the previously used NCP protocol, like working with different types of
networks and devices.

These and several other TCP functions allowed computer terminals to communicate and exchange information
on ARPANET.

Interface message processors

Interface message processors (IMPs) were important components of ARPANET, serving as intermediaries
between host computers and the network. Here’s how they worked:

• IMPs were specialized computers designed to connect the host computers and the ARPANET network.
Each host computer had a dedicated IMP responsible for transmitting and receiving data packets.

• The interface message processors acted as gateways. They translated data from the host computers into
packets to be transmitted across the network. IMPs also added the necessary information to data
packets and ensured they were correctly routed to their destinations.

• When a host computer wanted to send data, it would send it to its dedicated IMP. The IMP would break
the data into packets, add appropriate information (e.g., headers) and transmit them.

• The receiving interface message processor would strip off the headers and deliver the data to the host
computer.

Plutarch:

Plutarch is a network architecture designed to address the limitations of the traditional Internet model, which
is struggling to cope with the increasing diversity of network types and applications.

• Context Awareness:

o Dynamic Adaptation: Plutarch-based networks can adjust their behavior based on real-time
information about the network environment, user preferences, and application requirements.
 o Contextual Decision Making: Network decisions are made by considering relevant context
data, leading to optimized network performance.

• Interstitial Functions:

o Modular Network: These are additional functionalities that can be inserted into the network to
provide specific services or modify network behavior.

o Flexibility: They allow for customization and adaptation of the network to meet specific needs.

• Subsumption of Existing Protocols:

o Interoperability: Plutarch can incorporate existing network protocols, ensuring compatibility

with existing infrastructure.

o Gradual Adoption: This allows for a phased transition to the Plutarch architecture.

• Overlay Networks:

o Virtual Networks: Plutarch can create virtual networks on top of existing physical networks to
isolate traffic and provide specific services.

o Resource Optimization: Overlay networks can optimize resource utilization and improve
performance.

Traid:

TRIAD, a network architecture designed to address the limitations of the traditional Internet, revolves around a few core
concepts:

1. Name-Based Addressing

• Core principle: Uses human-readable names (like domain names) instead of numerical IP addresses to identify
network endpoints.

• Benefits: Easier to remember, manage, and troubleshoot.

• Implications: Requires a robust naming system and efficient name resolution mechanisms.

2. Integrated Naming and Routing

• Concept: Combines the functions of name resolution (DNS) and packet forwarding (routing) into a unified
system.

• Advantages: Reduces latency and improves efficiency by eliminating the need for separate DNS lookups.
 • Challenges: Increased complexity in network infrastructure.

3. Content-Centric Networking (CCN)

• Focus: Shifts the network's attention from endpoints to content itself.

• Benefits: Improved content delivery, caching, and replication.

• Implications: Requires efficient mechanisms for content identification, discovery, and distribution.

4. Incremental Deployability

• Approach: Allows for gradual integration of TRIAD into existing networks without disrupting services.

• Advantages: Reduces implementation challenges and risks.

• Considerations: Careful planning and phased rollout are essential.

TRIAD aims to create a more scalable, efficient, and flexible network architecture capable of handling the increasing
demands of modern applications and services.