Internet Routing COS 598A Jennifer Rexford http://www.cs.princeton.edu/~jrex/teaching/spring2005 Tuesdays/Thursdays 11:00am-12:20pm Who am I, and Who are You? • Who am I? – Joined the CS faculty in Feb 2005 (i.e., today) – Worked for 8.5 years at AT&T Labs—Research – Research on routing protocols, network measurement, and network operations • Who are you, and what do you do? – Introductions… What is Internet Routing? • The glue that holds the Internet together • How routers know where to forward packets • How operators control the load on their links • How networks achieve business relationships 4 3 5 2 1 Client 7 6 Web server What Does This Course Cover? • Internet architecture – Best-effort packet-delivery service – Intradomain and interdomain routing • Network topology – Inside a network, and between networks • Traffic engineering – Getting the traffic to go where you want • Convergence – Delay to respond to change – Whether the protocol ever converges What Does the Course Cover? (Continued) • Routers – Router hardware and software – Router configuration – Scaling to many destinations, routers, & networks • Measurement – Monitoring the routing protocols – Characterizing the routing system – Troubleshooting routing problems • Routing protocol security • New architectural directions Emphasis of the Course • Not so much on the protocols – …though we will cover BGP, OSPF, IS-IS, MPLS, and various other acronyms of the day • Or on the routers – …though we will talk about how routers work • But more on how people manage routing – Selecting which protocols to use – Deciding how to set the parameters – Troubleshooting problems as they arise – Preventing attacks –… Structure of the Course • Classroom time – Mixture of lecture and discussion of papers • Readings – Selected research papers and surveys – Videocasts of presentations (e.g., from NANOG) – Optional short “food for thought” reading each week • Course project – Literature survey, measurement or simulation study, protocol design, theoretical analysis, etc. • Grading – Final course project (written report and oral presentation) – Class participation (written reviews, class discussion, etc.) Today, and Thursday • Goal – Explain IP best-effort delivery model • Today – What is the service model? – How can you do anything useful with this? • Thursday – How do the routers support the service model? – How do the routing protocols work? IP Service Model: Best-Effort Packet Delivery • Packet switching – Send data in packets – Header with source & destination address • Best-effort delivery – Packets may be lost – Packets may be corrupted – Packets may be delivered out of order source destination IP network IP Service Model: Why Packets? • Data traffic is bursty – Logging in to remote machines – Exchanging e-mail messages • Don’t want to waste reserved bandwidth – No traffic exchanged during idle periods • Better to allow multiplexing – Different transfers share access to same links • Packets can be delivered by most anything – RFC 2549: IP over Avian Carriers (aka birds) • … still, packet switching can be inefficient – Extra header bits on every packet IP Packet Structure usually 20 bytes usually IPv4 4-bit 8-bit 4-bit Version Header Type of Service Length (TOS) 3-bit Flags 16-bit Identification fragments 8-bit Time to Live (TTL) 16-bit Total Length (Bytes) 8-bit Protocol 13-bit Fragment Offset 16-bit Header Checksum 20-byte Header 32-bit Source IP Address more later 32-bit Destination IP Address Options (if any) Payload error check header IP Service Model: Why Best-Effort? • It’s easier not to make promises – Don’t need to reserve bandwidth and memory – Don’t need to do error detection & correction – Don’t need to remember from one packet to next • Easier to survive failures – Transient disruptions are okay during failover • … but, applications do want efficient, accurate transfer of data in order, in a timely fashion IP Service Model: Best-Effort is Enough • No error detection or correction – Higher-level protocol can provide error checking • Successive packets may not follow the same path – Not a problem as long as packets reach the destination • Packets can be delivered out-of-order – Receiver can put packets back in order (if necessary) • Packets may be lost or arbitrarily delayed – Sender can send the packets again (if desired) • No network congestion control (beyond “drop”) – Sender can slow down in response to loss or delay Layering in the IP Protocols HTTP Telnet FTP DNS Transmission Control Protocol (TCP) User Datagram Protocol (UDP) Internet Protocol SONET Ethernet RTP ATM Transmission Control Protocol (TCP) • Communication service (socket) – Ordered, reliable byte stream – Simultaneous transmission in both directions • Key mechanisms at end hosts – – – – Retransmit lost and corrupted packets Discard duplicate packets and put packets in order Flow control to avoid overloading the receiver buffer Congestion control to adapt sending rate to network load TCP connection source network destination Source and Destination Port Numbers • Motivation for port numbers – Unique identifier of the TCP connection on each end – Necessary to (de)multiplex packets at the end-points • Assigning port numbers – Port numbers below 1024 are assigned – Well-known port numbers for common applications • Web client contacting a web server – – – – Browser click results in creation of a TCP socket Client machine assigns an available port (>=1024) Client machine requests a connection with the server Open TCP connection to port 80 at the server Opening and Closing a TCP Connection B A time • Three-way handshake to establish connection – Host A sends a SYN to the host B – Host B returns a SYN and acknowledgement – Host A sends an ACK to acknowledge the SYN ACK • Four-way handshake to close the connection – Finish (FIN) to close and receive remaining bytes , or – Reset (RST) to close and not receive remaining bytes Lost and Corrupted Packets • Detecting corrupted and lost packets – Error detection via checksum on header and data – Sender sends packet, sets timeout, and waits for ACK – Receiver sends ACKs for received packets – Sender infers loss from timeout or duplicate ACKs • Retransmission by sender – Sender retransmits lost/corrupted packets – Receiver reassembles and reorders packets – Receiver discards corrupted and duplicated packets TCP Flow and Congestion Control • Window-based flow control – Sender limits number of outstanding bytes (window size) – Receiver window ensures data does not overflow receiver • Adapting to network congestion congestion window – Congestion window tries to avoid overloading the network (increase with successful delivery, decrease with loss) – TCP connection starts with small initial congestion window congestion avoidance slow start time User Datagram Protocol (UDP) • Some applications do not want or need TCP – Avoid overhead of opening/closing a connection – Avoid recovery from lost/corrupted packets – Avoid sender adaptation to loss/congestion • Example applications that use UDP – Multimedia streaming applications – Domain Name System (DNS) queries/replies • Dealing with the growth in UDP traffic – Interference with TCP performance – Pressure to apply congestion control – Future routers may enforce “TCP-friendly” behavior Domain Name System (DNS) • Properties of DNS – Hierarchical name space divided into zones – Translation of names to/from IP addresses – Distributed over a collection of DNS servers • Client application – Extract server name (e.g., from the URL) – Invoke system call to trigger DNS resolver code – E.g., gethostbyname() on “www.foo.com” • Server application – Extract client IP address from socket – Optionally invoke system call to translate into name – E.g., gethostbyaddr() on “12.34.158.5” Domain Name System unnamed root com edu org generic domains bar uk ac zw arpa country domains ac inaddr west east cam 12 foo my usr 34 my.east.bar.edu usr.cam.ac.uk 56 12.34.56.0/24 DNS Resolver and Local DNS Server Root server 3 4 Application DNS cache 5 1 10 DNS resolver DNS query 2 6 Local DNS server Top-level domain server 7 DNS response 9 8 Second-level domain server Caching based on a time-to-live (TTL) assigned by the DNS server responsible for the host name to reduce latency in DNS translation. Application-Layer Protocols • Messages exchanged between applications – Syntax and semantics of the messages between hosts – Tailored to the specific application (e.g., Web, e-mail) – Messages transferred over transport connection (e.g., TCP) • Popular application-layer protocols – Telnet, FTP, SMTP, NNTP, HTTP, … GET /index.html HTTP/1.1 Client HTTP/1.1 200 OK Server Example: Many Steps in Web Download Browser cache DNS resolution TCP open 1st byte response Last byte response Sources of variability of delay • Browser cache hit/miss, need for cache revalidation • DNS cache hit/miss, multiple DNS servers, errors • Packet loss, high RTT, server accept queue • RTT, busy server, CPU overhead (e.g., CGI script) • Response size, receive buffer size, congestion • … downloading embedded image(s) on the page IP Suite: End Hosts vs. Routers host host HTTP message HTTP TCP segment TCP router IP Ethernet interface HTTP IP packet Ethernet interface IP TCP router IP packet SONET interface SONET interface IP IP packet Ethernet interface This course focuses on the routers… IP Ethernet interface Happy Routers Make Happy Packets • Routers forward packets – Forward incoming packet to outgoing link – Store packets in queues – Drop packets when necessary • Routers compute paths – Routers run routing protocols – Routers compute forwarding tables • A famous quotation from RFC 791 – “A name indicates what we seek. An address indicates where it is. A route indicates how we get there.” -- Jon Postel Reading for Thursday • Two classic papers – End-to-end arguments in system design (1984) – Design philosophy of the DARPA Internet protocols (1988) • New perspectives on success of the Internet – Tussle in cyberspace: Defining tomorrow’s Internet (2002) Backup Slides • TTL and traceroute Time-to-Live Field • Potential robustness problem – Routing loops can cause packets to cycle forever – Confusing if the packet arrives much later • Time-to-live field in packet header – TTL field decremented by each router on the path – Packet is discarded when TTL field reaches 0… – …and send “timer expired” message to source Traceroute: Measuring the Forwarding Path • Time-To-Live field in IP packet header – Source sends a packet with a TTL of n – Each router along the path decrements the TTL – “TTL exceeded” sent when TTL reaches 0 • Traceroute tool exploits this TTL behavior TTL=1 source Time exceeded destination TTL=2 Send packets with TTL=1, 2, 3, … and record source of “time exceeded” message