Computer Science 425 Distributed Systems

1
Computer Science 425Distributed Systems

• Lecture 12
• NETWORKING Part II

2
IP Packet Layout

• IP address allocation problem
• Network administrators may overestimate (or
  underestimate) future growth
• of their networks and request Class B (or Class
  C) address.
• Three existing solutions
• IP v6
• Classless inter-domain routing (CIDR)
• Problem scarcity of Class B addresses, and
  plenty of class C addresses
• Network Address Translation (NAT)

3
(1) IPv6 Header Layout
0-8 time-insensitive traffic 8-15
time-sensitive traffic
Establishment of virtual circults
IPv6 addresses are 128 bits (16 bytes) long
sufficient to provide 7 x 1023 IP addresses per
square meter across the entire surface of the
Earth.
4
(2) CIDR

• Classless Interdomain Routing
• Problem
• Shortage of class B addresses requires sites with
  multiple networks to obtain class C network IDs,
  instead of a single class B network ID
• Every class C network requires now a routing
  table entry
• Solution
• CIDR prevents this explosion of Internet routing
  tables
• Basic Concept
• Allocate multiple IP addresses that allow
  summarization into a smaller number of routing
  table entries
• Use Network Prefix and Mask in routing tables to
  allow for single entry for the range of class C
  addresses in the same subnet

5
(3) Network Address Translation (NAT)

• Not all devices have assigned globally-unique IP
  addresses
• NAT-enabled router translates global registered
  IP address into unregistered IP addres
• Example
• Home network has been allocated single registered
  IP address 83.215.152.95 by the Internet Service
  Provider
• All Internet-enabled devices on the home network
  have been assigned unregistered IP addresses on
  the 192.168.1.x Class C subnet
• Internal devices are allocated individual IP
  addresses dynamically via Dynamic Host
  Configuration Protocol (DHCP)
• It is conventional to use addresses from one of
  the three blocks of addresses (10.z.y.x,
  172.16.y.x, 192.168.y.x) that IANA considers for
  private internets

6
NAT Protocol

• Sender on local home network sends UDP or TCP
  packet to a host outside
• Router receives the packet and saves the source
  IP address and port number in its address
  translation table
• Router replaces the source address in the packet
  with routers IP address and the source port with
  a virtual port number that indexes the table slot
  containing senders address
• Packet is forwarded to the destination by the
  router
• When the router receives UDP or TCP packet, it
  uses the destination port number to access slot
  in the address translation table. It replaces the
  destination address and port with the internal
  address/port and forwards to the internal home
  network

7
PCs,routers, switches nodes
links edges
The Internet (Internet Mapping Project, color
coded by ISPs)
8
We have LANs what are these othernodes that
connect them?

• Hubs connects Ethernet hosts
• Simply repeats packet (connects multiple Ethernet
  devices together making them act as a single
  network segment
• Highest layer data link layer
• Bridges connect different types of networks
• Translate a packet from one format into another,
  e.g., Ethernet to FDDI
• Highest layer data link layer (multiple)
• Switches connect and route between two LANs
  (e.g., two Ethernet segments)
• Routing algorithms
• Highest layer data link layer
• Routers connect and route between two networks
• Routing algorithms
• Highest layer network layer
• Tunneling encapsulate and send an
  unimplemented protocols packets through from
  network layer to network layer
• IPv6 packets can be tunneled through the IPv4
  network

9
How do peer layers talk?
MAC/Data Link layer match the MAC addresses (48
bits) Network Layer e.g., IP match the IP
address (32 bits) Transport Layer e.g., TCP
match the port number (8 bits)
Application
Application message
Transport
port
TCP header
Network
TCP
IP header
MAC
Ethernet header
IP
(Physical)
Ethernet frame
10
Translation Who discovers these addresses for
the destination?

• Port number from the IP address ( a standard
  port number)
• IP address from the resource name/URL (through
  the DNS)
• Ethernet address from the IP address (through
  ARP)

11
ARP Address Resolution Protocol between IP and
Underlying Networks

• Most hosts are attached to a LAN by an interface
  board that only understands LAN addresses. For
  example, every Ethernet board is equipped with a
  globally unique 48-bit Ethernet address.
• The boards send and receive frames based on
  48-bit Ethernet addresses. They know nothing
  about the 32-bit IP addresses.
• Address Resolution Protocol (ARP) maps the IP
  addresses onto data link layer addresses (e.g.,
  Ethernet).

12
ARP Example
Routers have multiple network Interface
cards/devices. Each interface has a different
MAC/IP address.
13
ARP

• Suppose host 1s IP layer (192.31.65.7) gets a
  packet from its transport layer destined for
• 192.31.65.5 (host 2)?
• Host 1s IP layer broadcasts an ARP packet onto
  the Ethernet asking
• Who owns IP address 192.31.65.5?''
• Host 2 responds with its Ethernet address (E2).
• The IP layer on host 1 builds an Ethernet frame
  addressed to E2, puts the IP packet
• in the payload field, and transmits it on the
  Ethernet.
• The Ethernet board of host 2 detects this frame
  and causes an interrupt, to deliver the packet to
• the IP layer on host 2.
• Thus, the packet is transmitted from host 1s IP
  layer to host 2s IP layer

14
ARP

• The performance of ARP can be improved by caching
  the broadcast results.
• Host 1 can include its own IP to Ethernet mapping
  in the ARP packet.

15
ARP

• Suppose host 1s IP layer (192.31.65.7) gets a
  packet from its transport layer with
• destination address set to 192.31.63.8 (host 4)?
  
• Host 1s IP layer broadcasts an ARP packet onto
  the Ethernet asking
• Who owns IP address 192.31.63.8?''
• Router E3/F1 responds with its Ethernet address
  (E3).
• The IP layer on host 1 transmits an Ethernet
  frame addressed to E3
• The E3 Ethernet board of router F1 receives the
  frame and delivers it to the IP layer on F1
• F1s IP layer knows from the destination address
  of 192.31.63.8 in the packet that it has to
• be next sent to 192.31.60.7.
• F1s IP layer sends an ARP on the FDDI ring for
  IP address 192.31.60.7.
• Router E4/F3 replies.
• F1s IP layer transmits an FDDI frame
  (containing the packet) addressed to F3
• The FDDI board of F3 receives the frame, and
  delivers it to the IP layer.
• F3 knows from the destination address of
  192.31.63.8 in the packet that it has to
• be next sent out to 192.31.63.8 through the
  interface E4
• F3 does an ARP on the EE Ethernet for IP address
  192.31.63.8
• Host 4 responds with E6
• F3 transmits the packet inside an Ethernet frame
  addressed to E6, and it reaches host 4s IP layer

16
Knows How?

• Routing Algorithms!
• In the Network layer

17
Routing Algorithms

• Programmed in the network layer
• determine the next hop, given the destination
  IP address,
• thus determine the route for each packet as it
  travels through the net,
• dynamically update routing information to reflect
  failures, changes and congestion.
• Two approaches
• link-state (e.g., OSPF)
• Every node knows status of each link in the
  network
• distance-vector (e.g., RIP)
• Every node knows the next-hop for each possible
  destination LAN
• Information maintained as a table
• Tables updated either
• Proactively periodically, or
• Reactively when a neighbor/some link status
  changes

18
Distance Vector Routing

• Also termed as distributed Bellman-Ford algorithm
  or Ford-Fulkerson algorithm, included in RIP
  (routing information protocol), AppleTalk, and
  Cisco routers.
• Each node/router maintains a table indexed by
  each destination node. Entry gives the best known
  distance to destination and which link to use for
  forwarding.
• Once every T seconds each router sends to each
  neighbor its own entire table (proactive)

19
Distance Vector Routing
To Link Cost A 1
1 C 2 1 D
4 2 E 4
1 B local

Routing Table for A
Routing Table for B
To Link Cost B 1
1 C 1 2 D
3 1 E 1
2 A local
To Link Cost A 2
2 B 2 1 D
5 2 E 5
1 C local
Link number (all links have cost1)
Routing Table for C
20
Pseudo-Code for RIP
Send Each t seconds or when Tl changes, send Tl
on each non-faulty outgoing link. Receive
Whenever a routing table Tr is received on link
n for all rows Rr in Tr if (Rr.link not equal
n) Rr.cost Rr.cost 1 Rr.link n if
(Rr.destination is not in Tl) add Rr to Tl
// add new destination to Tl else for all rows Rl
in Tl if (Rr.destination Rl.destination and
(Rr.cost lt Rl.cost or Rl.link n)) Rl Rr //
Rr.cost lt Rl.cost remote node has better
route // Rl.link n remote node is more
authoritative
21
Link State Routing

• Each router must
• Discover its neighbors and learn their network
  addresses
• When a router is booted, it learns who its
  neighbors are by sending a special Hello packet
  on each point-to-point link.
• The router on the other end sends back a reply.
• Measure the delay or cost to each of its
  neighbors
• A router sends a special Echo packet over the
  link that the other end sends back immediately.
  By measuring the round-trip time, the sending
  router gets a reasonable delay estimate.
• Construct a packet telling all it has just
  learned.
• Broadcast this packet

22
Link State Routing

• A router broadcasts a link-state-advertisement
  (LSA) packet after booting, as well as
  periodically (or upon topology change). Packet
  forwarded only once, TTL-restricted
• Initial TTL is very high.

23
Link State Routing

• Broadcast the LSA packet to all other routers in
  the subnet.
• Each packet contains a sequence number that is
  incremented for each new LSA packet sent.
• Each router keeps track of all the (source
  router, sequence) pairs it sees. When a new LSA
  packet comes in, it is checked against the pairs.
  If the received packet is new, it is forwarded
  on all the links except the one it arrived on.
• The age of each packet is included and is
  decremented once per time unit. When the age hits
  zero, the information is discarded. Initial age
  very high
• For routing a packet, since the source knows the
  entire network graph, it simply computes the
  shortest path (actual sequence of nodes) locally
  using the Dijkstras algorithm.

24
Transport Layer Transmission Control
Protocol

• Function 0 provide an application with a
  connection-oriented view of the network (IP is
  connectionless)
• Function 1 (Message decomposition and
  reassembly) Breaks messages into packets at the
  transmitting end and reassembles packets into
  messages at the receiving end.
• E.g., using identification and fragment offset
  fields in IPv4 header
• Function 2 (Multiplexing and demultiplexing)
  Multiplexes several lower-rate sessions, all from
  the same source and all going to the same
  destination, into one session at the network
  layer.
• Function 3 (Reliable communication) Provides
  reliability to the application by acks
  retransmissions in an end to end manner
• Function 4 (End-to-end congestion/flow control)
  Reduces rate at which data is sent when
  congestion is detected in the network. (TCP
  friendliness)
• All these functionalities are a part of TCP.

25
Session/Presentation/Application Layers

• The session and presentation layers deal with
  setting up virtual sessions (e.g., SSL), data
  encryption, data compression etc.
• The application layer actually does the work
  required by the users, e.g., FTP, HTTP, p2p.
• The application/presentation/session layers are
  not clearly distinguished in the Internet
  protocol stack.

26
Next

• Tuesday (7 October) Midterm Exam.
• Location Here (Classroom).
• Closed book.
• Duration 75 minutes.
• 2.00 pm - 3.15 pm

27
Midterm Review

• Topics Lectures 1-10
• What is a distributed system and examples of
  distributed systems
• Time and Synchronization
• Cristians Algorithm, Berkeley Algorithm and NTP
• Happens-Before Relation, Lamport Timestamps,
  Vector Logical Clocks
• Global States and Snapshots
• Consistent state, liveness, safety properties
• Chandy Lamport Snapshot Algorithm
• Causality violation
• Multicast
• B-multicast
• Reliable multicast (integrity, validity,
  agreement properties)
• Ordered Multicast Total, FIFO, Causal Ordering

28
Midterm Review Topics

• Implementing FIFO Ordering, Total Ordering,
  Causal Ordering
• Mutual Exclusion
• Evaluation criteria (client delay,
  synchronization delay)
• Centralized control , token ring, Ricart and
  Agrawala, Maekawas Algorithms, Raymonds
  token-based approach
• Election
• Ring election, modified ring election, Bully
  algorithm
• Consensus problem
• Consensus problem in synchronous systems
• Consensus problem in asynchronous systems,
  commutative schedules
• Failure Detectors
• Properties of failure detector (completeness,
  accuracy)
• Ping-ack, heart-beating protocols
• Failure detection in distributed system (ring
  heart-beating,)
• Failure types

29
Midterm Review Topics

• Peer-to-peer systems
• Napster protocol and search
• Gnutella protocol and search
• FastTrack protocol and search
• Chord, distributed hash tables, search under peer
  failure, new peers joining