Upgrade to remove ads
Only $35.99/year

Ch 5 Network Management - Network and Transport Layers

Terms in this set (33)

How many ports are there?
-There are a total of 65535 ports used for connections

-Ports 0-1024 are well-known ports
What is TCP/IP
Transmission Control Protocol and Internet Protocol

-Originally developed as a single internet working protocol by Vint Cerf and Bob Kahn in 1974

-Later divided into the TCP and IP protocols

-Most common protocols of the Internet and in LANs, WANs, and backbone networks
TCP
-Most common transport layer protocol
-PDU (Protocol Data Unit) called a segment
-Used for reliable transmission of data
-160 -192 bits (20 -24 bytes) of overhead
User Datagram Protocol (UDP)
-Operates at the transport layer
-PDU called a segment
-Used in time-sensitive situations, for control messages, or when reliability is handled by the application layer
-32-64 bits (4-8 bytes) of overhead
IP version 4 (IPv4)
Internet Protocol Version 4
•Most common version of IP used
•32-bit addresses (2^32or ~4.29 billion possible)
•Exhaustion of address space
IP Version 6 (IPv6)
Internet Protocol Version 6
•128-bit addresses (2128or ~3.4 ×1038possible)
•Slowly being adopted due to IPv4 exhaustion
IPv4 Packet
-160-192 bits (20-24 bytes) of overhead
-Options field rarely used
IPv6 Packet
Fixed Header
-320 bits (40 bytes) of overhead
Linking to the application layer
-TCP/UDP may serve multiple application layer protocols
-Ports used to identify application (2-byte numbers)
-Many source/destination ports follow standards-Common port standards
•HTTP: TCP port 80
•HTTPS: TCP port 443
Segmenting
-Breaking up large files into smaller segments (and putting them back together)
-Segments may be passed individually to application layer or after reassembly
Session management
-A session can be thought of as a conversation between two computers or creating a virtual circuit
-Using a session to send data is also called connection-oriented messaging (TCP)
-Sending messages without establishing a session is connection-less messaging (UDP)
-TCP connections are opened using a three-way handshake
•SYN
•SYN-ACK
•ACK
-Sessions provide reliable end-to-end connections
Addressing
-Used to direct messages from source to destination
-Addresses are assigned in various ways (e.g., by system administrators, ICANN, hardware vendors, etc.)
-Addresses exist at different layers
-Addresses may be translated (resolved) from one layer to another (e.g., DNS, ARP)
Most common way to write addressing?
dot-decimal notation
•Easier for people to read and remember
•Breaks the address into four bytes and writes each byte in decimal notation instead of binary
•Example: 129.79.78.193
Classless Inter-Domain Routing (CIDR)
Uses subnet masks to more flexibly divide address space into subnets
-IP address: 129.79.78.193
-Subnet Mask: 255.255.255.0
Classful addressing
•Uses the first bits to determine number of hosts
•Discontinued, but nomenclature still used
Dynamic addressing
-Configuring each device manually is time consuming
-Assigning addresses permanently can be inefficient when devices are not connected to network
-A server can supply IP addresses automatically
Dynamic Host Configuration Protocol (DHCP)
•Most common protocol for dynamic addressing •Device sends out broadcast message
•DHCP responds with IP settings
•Addresses are "leased" for a length of time
Host (server) name resolution
•Translate host name to IP address
•e.g., www.indiana.edu → 129.79.78.193
•Domain Name Service (DNS)
MAC address resolution
•Identify MAC address of the next device in the circuit
•Address Resolution Protocol (ARP)
Copyright
Routing
-Process of identifying what path to have a packet take through a network from sender to receiver
-Routing Tables
•Used to make routing decisions
•Shows which path to send packets on to reach a given destination
•Kept by computers making routing decisions
-Routers
•Special purpose devices used to handle routing decisions on the Internet
•Maintain their own routing tables
Centralized Routing
-Routing decisions made by one computer
-Not common anymore
Decentralized Routing
-Decisions made by each node independently of one another
-Information needs to be exchanged to prepare routing tables
-Used by the Internet
Static Routing
-Fixed routing tables
-Manually configured by network managers
-Local adjustments when computers added or removed
Dynamic Routing
-Routing tables updated periodically
-Routers exchange information using protocols to update tables
Distance vector:
Dynamic Routing Algorithm based on the number of "hops" between two devices
Link state:
Dynamic Routing Algorithm based on the number of hops, circuit speed, and traffic congestion
•Provides more reliable, up to date paths to destinations
Routing Information Protocol (RIP)
-Dynamic distance vector protocol used for interior routing
-Operation•Network manager builds the routing table
•Routing tables broadcast periodically (e.g., every minute or so)
•When new computers are added, router counts "hops" and selects the shortest route
-Useful in smaller, less complex networks
Open Shortest Path First (OSPF)
-Dynamic link state protocol used for interior routing
-Most widely used interior routing protocol on large enterprise networks
-More reliable paths
-Less burdensome to the network because only updates sent
Enhanced Interior Gateway Routing Protocol (EIGRP)
-A dynamic link state protocol (developed by Cisco)
-Records transmission capacity, delay time, reliability and load for all paths
-Keeps the routing tables for its neighbors and uses this information in its routing decisions as well
Border Gateway Protocol (BGP)
-Dynamic distance vector protocol used for exterior routing
-Far more complex than interior routing protocols
-Provide routing info only on selected routes (e.g., preferred or best route)
Unicast
one computer to another computer
Broadcast
one computer to all computers in the network
Multicast
one computer to a group of computers (e.g., videoconference)
Sets with similar terms