Internetworking
15-‐213 / 18-‐213: Introduc2on to Computer Systems
20th Lecture, Nov. 5, 2013 Instructors:
Randy Bryant, Dave O’Hallaron, and Greg Kesden
2
A Client-‐Server Transac8on
Client
process Server
process 1. Client sends request
2. Server handles request 3. Server sends response
4. Client handles response
Resource
¢
Most network applica8ons are based on the client-‐server model:
§
A server process and one or more client processes§
Server manages some resource§
Server provides service by manipula2ng resource for clients§
Server ac2vated by request from client (vending machine analogy)Note: clients and servers are processes running on hosts (can be the same or different hosts)
Hardware Organiza8on of a Network Host
main memory I/O
bridge MI
ALU register file
CPU chip
system bus memory bus
disk controller graphics
adapter USB
controller
mouse keyboard monitor
disk I/O bus
Expansion slots
network adapter
network
4
Computer Networks
¢
A network is a hierarchical system of boxes and wires organized by geographical proximity
§
SAN (System Area Network) spans cluster or machine room§ Switched Ethernet, Quadrics QSW, …
§
LAN (Local Area Network) spans a building or campus§ Ethernet is most prominent example
§
WAN (Wide Area Network) spans country or world§ Typically high-‐speed point-‐to-‐point phone lines
¢
An internetwork (internet) is an interconnected set of networks
§
The Global IP Internet (uppercase “I”) is the most famous example of an internet (lowercase “i”)¢
Let’s see how an internet is built from the ground up
Lowest Level: Ethernet Segment
¢
Ethernet segment consists of a collec8on of hosts connected by wires (twisted pairs) to a hub
¢
Spans room or floor in a building
¢
Opera8on
§ Each Ethernet adapter has a unique 48-‐bit address (MAC address)
§ E.g., 00:16:ea:e3:54:e6
§ Hosts send bits to any other host in chunks called frames
§ Hub slavishly copies each bit from each port to every other port
§ Every host sees every bit
§ Note: Hubs are on their way out. Bridges (switches, routers) became cheap enough
host host host hub 100 Mb/s 100 Mb/s
port
6
Next Level: Bridged Ethernet Segment
¢
Spans building or campus
¢
Bridges cleverly learn which hosts are reachable from which ports and then selec8vely copy frames from port to port
host host host host host
hub 100 Mb/s bridge 100 Mb/s hub
host host
hub 100 Mb/s 100 Mb/s 1 Gb/s
host host host bridge
host host
hub
A B
C X
Y
Conceptual View of LANs
¢
For simplicity, hubs, bridges, and wires are oZen shown as a collec8on of hosts a[ached to a single wire:
host host ... host
8
Next Level: internets
¢
Mul8ple incompa8ble LANs can be physically connected by specialized computers called routers
¢
The connected networks are called an internet
host host ... host host host ... host
WAN WAN
LAN 1 and LAN 2 might be completely different, totally incompaHble (e.g., Ethernet, Fibre Channel, 802.11*, T1-‐links, DSL, …)
router router router
LAN 1 LAN 2
Logical Structure of an internet
¢
Ad hoc interconnec8on of networks
§
No par2cular topology§
Vastly different router & link capaci2es¢
Send packets from source to des8na8on by hopping through networks
§
Router forms bridge from one network to anotherrouter
router
router router
router router
host host
10
The No8on of an internet Protocol
¢
How is it possible to send bits across incompa8ble LANs and WANs?
¢
Solu8on:
§
protocol sodware running on each host and router§
smooths out the differences between the different networks¢
Implements an internet protocol (i.e., set of rules)
§
governs how hosts and routers should cooperate when they transfer data from network to network§
TCP/IP is the protocol for the global IP InternetWhat Does an internet Protocol Do?
¢
Provides a naming scheme
§
An internet protocol defines a uniform format for host addresses§
Each host (and router) is assigned at least one of these internet addresses that uniquely iden2fies it¢
Provides a delivery mechanism
§
An internet protocol defines a standard transfer unit (packet)§
Packet consists of header and payload§ Header: contains info such as packet size, source and des2na2on addresses
§ Payload: contains data bits sent from source host
12
LAN2
Transferring Data Over an internet
protocol soZware
client
LAN1 adapter
Host A
LAN1
data (1)
data PH FH1 (4)
data PH FH2 (6)
data (8)
data PH FH2 (5) LAN2 frame
protocol soZware LAN1
adapter LAN2 adapter
Router
data PH
(3) FH1
data PH FH1 (2)
internet packet
LAN1 frame
(7) data PH FH2
protocol soZware server
LAN2 adapter
Host B
PH: Internet packet header FH: LAN frame header
Other Issues
¢
We are glossing over a number of important ques8ons:
§
What if different networks have different maximum frame sizes?(segmenta2on)
§
How do routers know where to forward frames?§
How are routers informed when the network topology changes?§
What if packets get lost?¢
These (and other) ques8ons are addressed by the area of
systems known as computer networking
14
Global IP Internet
¢
Most famous example of an internet
¢
Based on the TCP/IP protocol family
§
IP (Internet protocol) :§ Provides basic naming scheme and unreliable delivery capability of packets (datagrams) from host-‐to-‐host
§
UDP (Unreliable Datagram Protocol)§ Uses IP to provide unreliable datagram delivery from process-‐to-‐process
§
TCP (Transmission Control Protocol)§ Uses IP to provide reliable byte streams from process-‐to-‐process over connec2ons
¢
Accessed via a mix of Unix file I/O and func8ons from the
sockets interface
Hardware and SoZware Organiza8on of an Internet Applica8on
TCP/IP Client
Network adapter
Global IP Internet
TCP/IP Server
Network adapter
Internet client host Internet server host
Sockets interface (system calls)
Hardware interface (interrupts)
User code
Kernel code Hardware and firmware
16
Basic Internet Components
¢
Internet backbone:
§
collec2on of routers (na2onwide or worldwide) connected by high-‐speed point-‐to-‐point networks¢
Internet Exchange Points (IXP):
§
router that connects mul2ple backbones (oden referred to as peers)§
Also called Network Access Points (NAP)¢
Regional networks:
§
smaller backbones that cover smaller geographical areas (e.g., ci2es or states)¢
Point of presence (POP):
§
machine that is connected to the Internet¢
Internet Service Providers (ISPs):
§
provide dial-‐up or direct access to POPsInternet Connec8on Hierarchy
IXP IXP
Backbone Backbone Backbone Backbone IXP
POP POP POP
Regional net
POP POP POP
POP POP
Small Business
Big Business ISP
POP
POP POP POP
Pgh employee
Cable modem
DC employee POP
T3
T1
ISP (for individuals) POP
DSL T1
ColocaHon sites Private
“peering”
agreements between two backbone
companies o[en bypass
IXP
18
A Programmer’s View of the Internet
¢
Hosts are mapped to a set of 32-‐bit IP addresses
§
128.2.203.179¢
The set of IP addresses is mapped to a set of iden8fiers called Internet domain names
§
128.2.203.179 is mapped to www.cs.cmu.edu¢
A process on one Internet host can communicate with a
process on another Internet host over a connecHon
IP Addresses
¢
32-‐bit IP addresses are stored in an IP address struct
§
IP addresses are always stored in memory in network byte order (big-‐endian byte order)§
True in general for any integer transferred in a packet header from one machine to another.§ E.g., the port number used to iden2fy an Internet connec2on.
/* Internet address structure */
struct in_addr {
unsigned int s_addr; /* network byte order (big-endian) */
};
Useful network byte-‐order conversion func8ons (“l” = 32 bits, “s” = 16 bits)
htonl: convert uint32_t from host to network byte order htons: convert uint16_t from host to network byte order ntohl: convert uint32_t from network to host byte order ntohs: convert uint16_t from network to host byte order
20
Do[ed Decimal Nota8on
¢
By conven8on, each byte in a 32-‐bit IP address is represented by its decimal value and separated by a period
§ IP address: 0x8002C2F2 = 128.2.194.242
¢
Func8ons for conver8ng between binary IP addresses and do[ed decimal strings:
§
inet_aton: domed decimal string → IP address in network byte order§
inet_ntoa: IP address in network byte order → domed decimal string§
“n” denotes network representa2on§
“a” denotes applica2on representa2onIP Address Structure
¢
IP (V4) Address space divided into classes:
¢
Network ID Wri[en in form w.x.y.z/n
§
n = number of bits in host address§
E.g., CMU wrimen as 128.2.0.0/16§ Class B address
¢
Unrouted (private) IP addresses:
10.0.0.0/8 172.16.0.0/12 192.168.0.0/16
Class A Class B Class C Class D Class E
0 1 2 3 8 16 24 31 0 Net ID Host ID
Host ID
Host ID Net ID
Net ID
Mul2cast address
Reserved for experiments 1 0
1 0 1
1 1 0 1 1 1 1 1
22
Internet Domain Names
.net .edu .gov .com cmu berkeley
mit
cs ece
greatwhite
128.2.220.10
ics
unnamed root
sp
i386-‐f7
128.2.200.47
amazon
www
2007.171.166.252
First-‐level domain names
Second-‐level domain names
Third-‐level domain names
Domain Naming System (DNS)
¢
The Internet maintains a mapping between IP addresses and domain names in a huge worldwide distributed database called DNS
§
Conceptually, programmers can view the DNS database as a collec2on of millions of host entry structures:¢
Func8ons for retrieving host entries from DNS:
§
gethostbyname: query key is a DNS domain name.§
gethostbyaddr: query key is an IP address./* DNS host entry structure */
struct hostent {
char *h_name; /* official domain name of host */
char **h_aliases; /* null-terminated array of domain names */
int h_addrtype; /* host address type (AF_INET) */
int h_length; /* length of an address, in bytes */
char **h_addr_list; /* null-terminated array of in_addr structs
*/
};
24
Proper8es of DNS Host Entries
¢
Each host entry is an equivalence class of domain names and IP addresses
¢
Each host has a locally defined domain name localhost which always maps to the loopback address 127.0.0.1
¢
Different kinds of mappings are possible:
§
Simple case: one-‐to-‐one mapping between domain name and IP address:§ greatwhile.ics.cs.cmu.edu maps to 128.2.220.10
§
Mul2ple domain names mapped to the same IP address:§ eecs.mit.edu and cs.mit.edu both map to 18.62.1.6
§
Mul2ple domain names mapped to mul2ple IP addresses:§ google.com maps to mul2ple IP addresses
§
Some valid domain names don’t map to any IP address:§ for example: ics.cs.cmu.edu
A Program That Queries DNS
int main(int argc, char **argv) { /* argv[1] is a domain name */
char **pp; /* or dotted decimal IP addr */
struct in_addr addr;
struct hostent *hostp;
if (inet_aton(argv[1], &addr) != 0)
hostp = Gethostbyaddr((const char *)&addr, sizeof(addr), AF_INET);
else
hostp = Gethostbyname(argv[1]);
printf("official hostname: %s\n", hostp->h_name);
for (pp = hostp->h_aliases; *pp != NULL; pp++) printf("alias: %s\n", *pp);
for (pp = hostp->h_addr_list; *pp != NULL; pp++) { addr.s_addr = ((struct in_addr *)*pp)->s_addr;
printf("address: %s\n", inet_ntoa(addr));
} }
26
Using DNS Program
linux> ./dns greatwhite.ics.cs.cmu.edu
official hostname: greatwhite.ics.cs.cmu.edu address: 128.2.220.10
linux> ./dns 128.2.220.11
official hostname: ANGELSHARK.ICS.CS.CMU.EDU address: 128.2.220.11
linux> ./dns www.google.com
official hostname: www.google.com address: 74.125.131.103
address: 74.125.131.104 address: 74.125.131.105 address: 74.125.131.106 address: 74.125.131.147 address: 74.125.131.99
Querying DIG
¢
Domain Informa8on Groper (dig) provides a scriptable command line interface to DNS
linux> dig +short greatwhite.ics.cs.cmu.edu 128.2.220.10
linux> dig +short -x 128.2.220.11 ANGELSHARK.ICS.CS.CMU.EDU.
linux> dig +short google.com 72.14.204.104
72.14.204.147 72.14.204.99 72.14.204.103
28
Internet Connec8ons
¢
Clients and servers communicate by sending streams of bytes over connecHons:
§
Point-‐to-‐point, full-‐duplex (2-‐way communica2on), and reliable.¢
A socket is an endpoint of a connec8on
§
Socket address is an IPaddress:port pair¢
A port is a 16-‐bit integer that iden8fies a process:
§
Ephemeral port: Assigned automa2cally on client when client makes a connec2on request§
Well-‐known port: Associated with some service provided by a server (e.g., port 80 is associated with Web servers)¢
A connec8on is uniquely iden8fied by the socket addresses of its endpoints (socket pair)
§
(cliaddr:cliport, servaddr:servport)Pumng it all Together:
Anatomy of an Internet Connec8on
Connec8on socket pair
(128.2.194.242:51213, 208.216.181.15:80)
Server (port 80) Client
Client socket address
128.2.194.242:51213 Server socket address 208.216.181.15:80
Client host address
128.2.194.242 Server host address
208.216.181.15
30
Evolu8on of Internet
¢
Original Idea
§
Every node on Internet would have unique IP address§ Everyone would be able to talk directly to everyone
§
No secrecy or authen2ca2on§ Messages visible to routers and hosts on same LAN
§ Possible to forge source field in packet header
¢
Shortcomings
§
There aren't enough IP addresses available§
Don't want everyone to have access or knowledge of all other hosts§
Security issues mandate secrecy & authen2ca2onEvolu8on of Internet: Naming
¢
Dynamic address assignment
§
Most hosts don't need to have known address§ Only those func2oning as servers
§
DHCP (Dynamic Host Configura2on Protocol)§ Local ISP assigns address for temporary use
¢
Example:
§
Laptop at CMU (wired connec2on)§ IP address 128.2.213.29 (bryant-tp4.cs.cmu.edu)
§ Assigned sta2cally
§
Laptop at home§ IP address 192.168.1.5
§ Only valid within home network
32
Evolu8on of Internet: Firewalls
¢
Firewalls
§
Hides organiza2ons nodes from rest of Internet§
Use local IP addresses within organiza2on§
For external service, provides proxy service1. Client request: src=10.2.2.2, dest=216.99.99.99
2. Firewall forwards: src=176.3.3.3, dest=216.99.99.99
3. Server responds: src=216.99.99.99, dest=176.3.3.3
4. Firewall forwards response: src=216.99.99.99, dest=10.2.2.2
Corpora8on X
Firewall
Internet
10.2.2.2 1
4 2
3 176.3.3.3
216.99.99.99
Next Time
¢
How to use the sockets interface to establish Internet connec8ons between clients and servers
¢