Understanding ATM and IP Switching: Enhancing Network Performance

1. Data-link layer Specific link layers and devices CSE524: Lecture 17

2. Where we’re at… • Internet architecture and history • Internet protocols in practice • Application layer • Transport layer • Network layer • Data-link layer • Functions • Specific link layer examples and devices • Physical layer

3. DL/NL: ATM • ATM • Replace existing Internet protocols with a more “robust” architecture • Network architecture to support • Multiple service classes and per-flow guarantees • Virtual circuits to support real-time applications • Explicit rate signaling and resource allocation • Covered as a data-link layer…

4. Internet “elastic” datagram service, no strict timing req. Computer communication only “smart” end systems (computers) can adapt, perform control, error recovery simple inside network, complexity at “edge” many link types different characteristics uniform service difficult ATM evolved from telephony, strict timing and reliability requirements Computer and human communication need for guaranteed service “dumb” end systems telephones complexity inside network DL/NL: Internet vs. ATM

5. DL/NL: ATM Layer: Virtual Circuits • VC transport: cells carried on VC from source to dest • call setup, teardown for each call before data can flow • each packet carries VC identifier (not destination ID) • every switch on source-dest path maintain “state” for each passing connection • link, switch resources (bandwidth, buffers) may be allocated to VC: to get circuit-like perf. • Permanent VCs (PVCs) • long lasting connections • typically: “permanent” route between to IP routers • Switched VCs (SVC): • dynamically set up on per-call basis

6. DL/NL: ATM VCs • Advantages of ATM VC approach: • QoS performance guarantee for connection mapped to VC (bandwidth, delay, delay jitter) • Drawbacks of ATM VC approach: • Overhead in call setup for SVCs • SVC introduces call setup latency, processing overhead for short lived connections • Lack of scalability for PVCs • one PVC between each source/dest pair does not scale (N*2 connections needed)

7. DL/NL: ATM Layer: ATM cell Cell header Cell format • 5-byte ATM cell header • 48-byte payload (fixed) • Why?: small payload -> short cell-creation delay for digitized voice • halfway between 32 and 64 (compromise!)

8. DL/NL: ATM cell header • VCI: virtual channel ID • will change from link to link thru net • PT:Payload type (e.g. RM cell versus data cell) • CLP: Cell Loss Priority bit • CLP = 1 implies low priority cell, can be discarded if congestion • HEC: Header Error Checksum • cyclic redundancy check

9. DL: ATM: network or link layer? • Vision • ATM end-to-end from desktop to desktop • Both a network and a data-link layer technology • Reality • Used mostly as a switched link-layer to connect IP routers • “IP over ATM” • replace IP network+routers with ATM network+switches • At edges, map ATM addresses to IP addresses and vice-versa

10. DL: ATM and “IP switching” • ATM advantages • Lookup of VCID = O(1), Lookup of IP routes O(log n) • One-time route lookup and circuit establishment, all subsequent traffic switched • ATM disadvantages • Complex signaling and routing for establishing communication • Difficulty in mapping IP traffic dynamically onto ATM circuits • Goal • Maintain IP infrastructure • Accelerate it with labels to support O(1) lookups a la ATM • Solution • Ipsilon and “IP switching” • http://pnewman.org/papers/infocom96.pdf

11. IP over ATM versus IP switching ATM network control ATM label switching IP network control IP routing IP network control IP routing IP network control ATM label switching IP network control IP routing IP network control IP routing

12. DL: ATM and “IP switching” • In a nutshell • Start with ATM switch • Rip out ATM signaling and routing • Add IP routing software • Add Flow classifier to map unknown flows to underlying ATM virtual circuit ID • Attach VCID and allow downstream nodes to do the same • Operation • Upon arrival of first packet in flow • Record unknown incoming VCID • Lookup IP flow and map it to an outgoing virtual circuit ID (label) using IP routing software • Create incomingVCID to outgoingVCID table entry for subsequent packets • Subsequent packets • Switched in hardware using VCID after flow classified at edge • IP packet forwarding done as label index lookup O(1) versus IP route lookup O(log n)

13. DL: ATM and “IP switching” • Later generalized as MPLS (multi-protocol label switching) • “Layer 2 ½” • Not tied to ATM • Extensible to IPv6 • Half-way in between data-link addresses and IP addresses • Labeling done within a cloud versus link-local (data-link addresses) and global (IP addresses) • http://www.rfc-editor.org/rfc/rfc3031.txt • Used as a tool for traffic engineering • http://www.rfc-editor.org/rfc/rfc2702.txt

14. DL: X.25 and Frame Relay Like ATM: • wide area network technologies • virtual circuit oriented • origins in telephony world • Not really a link layer but.... • Viewed as link layers by IP protocol • Used mostly to carry IP datagrams between IP routers • Going the way of the dinosaurs....

15. DL: X.25 • X.25 builds VC between source and destination for each user connection • Per-hop control along path • error control (with retransmissions) on each hop using LAP-B • variant of the HDLC protocol • developed when bit error rates over long-haul copper links were orders of magnitude higher • per-hop flow control using credits • congestion arising at intermediate node propagates to previous node on path • back to source via back pressure

16. DL: IP versus X.25 • X.25: reliable in-sequence end-end delivery from end-to-end • “intelligence in the network” • built for dumb terminals accessing mainframes • IP: unreliable, out-of-sequence end-end delivery • “intelligence in the endpoints” • 2000 • gigabit routers: limited processing possible • CPU capacity at end-hosts • IP wins

17. DL: Frame Relay • Designed in late ‘80s, widely deployed in the ‘90s • Second-generation X.25 • Frame relay service: • no error control • no flow control • End-to-end congestion control • Some QoS mechanisms

18. DL: Frame Relay (more) • Designed to interconnect corporate customer LANs • typically permanent VC’s: “pipe” carrying aggregate traffic between two routers • switched VC’s: as in ATM • corporate customer leases FR service from public Frame Relay network (eg, Sprint, ATT)

19. DL: Frame Relay (more) address data CRC flags flags • Flag bits, 01111110, delimit frame • address: • 10 bit VC ID field • 3 congestion control bits • FECN: forward explicit congestion notification (frame experienced congestion on path) • BECN: congestion on reverse path • DE: discard eligibility • Precursor to IP DiffServ and ECN

20. DL: Frame Relay -VC Rate Control • Committed Information Rate (CIR) • defined, “guaranteed” for each VC • negotiated at VC set up time • customer pays based on CIR • DE bit: Discard Eligibility bit • Edge FR switch measures traffic rate for each VC; marks DE bit • DE = 0: high priority, rate compliant frame; deliver at “all costs” • DE = 1: low priority, eligible for discard when congestion • Precursor to IP DiffServ • Can be used to support higher layer QoS mechanisms

21. DL: Link-layer devices Q: Why not just one big LAN? • Limited amount of supportable traffic: on single LAN, all stations must share bandwidth • limited length: 802.3 specifies maximum cable length • large “collision domain” (can collide with many stations) • limited number of stations: 802.5 have token passing delays at each station

22. DL: Hubs • Effectively a physical layer device • Multi-port repeater • Repeater operating at bit level • Repeat received bits on one interface to all other interfaces • Hubs can be arranged in a hierarchy (or multi-tier design), with backbone hub at its top

23. DL: Hubs (more) • Each connected LAN referred to as LAN segment • Hubs do not isolate collision domains: node may collide with any node residing at any segment in LAN • Hub Advantages: • simple, inexpensive device • Multi-tier provides graceful degradation: portions of the LAN continue to operate if one hub malfunctions • extends maximum distance between node pairs

24. DL: Hub limitations • single collision domain results in no increase in max throughput • multi-tier throughput same as single segment throughput • individual LAN restrictions pose limits on • number of nodes in same collision domain • total allowed geographical coverage • cannot connect different Ethernet types (e.g., 10BaseT and 100baseT)

25. DL: Bridges • Link Layer devices: operate on Ethernet frames, examining frame header and selectively forwarding frame based on its destination • Bridge isolates collision domains since it buffers frames • When frame is to be forwarded on segment, bridge uses CSMA/CD to access segment and transmit

26. DL: Bridges (more) • Bridge advantages: • Isolates collision domains resulting in higher total max throughput, and does not limit the number of nodes nor geographical coverage • Can connect different type Ethernet since it is a store and forward device • Transparent: no need for any change to hosts LAN adapters

27. DL: Backbone Bridge

28. DL: Interconnection Without Backbone • Not recommended for two reasons: • - single point of failure at Computer Science hub • - all traffic between EE and SE must path over CS segment

29. DL: Bridges: frame filtering, forwarding • Bridges filter packets • same-LAN -segment frames not forwarded onto other LAN segments • Forwarding: • how to know which LAN segment on which to forward frame? • looks like a routing problem • Solution: Learning bridges • Monitor traffic to build a cache of which nodes are downstream of which ports • Selectively forward frames based on cache entries • Flood network for frames with unknown (MAC) destinations

30. DL: Bridge Filtering • Bridges maintain filtering tables • Indicate which hosts can be reached through which interfaces • When frame received, bridge “learns” location of sender • Records sender port location in filtering table • Filtering table entry: • (Node LAN Address, Bridge Interface, Time Stamp) • Stale entries in Filtering Table dropped (TTL can be 60 minutes)

31. DL: Bridge Filtering • filtering procedure: ifdestination is on LAN on which frame was received then drop the frame else{ lookup filtering table if entry found for destination then forward the frame on interface indicated; else flood; /* forward on all but the interface on which the frame arrived*/ }

32. DL: Bridge Learning: example • C sends frame to D • Bridge has no info about D • Bridge notes that C is on LAN segment #1 • Bridge floods to both LAN segments #2 and #3 • frame ignored on upper LAN • frame received by D • D replies back with frame to C • Bridge knows C is on LAN segment #1 • Bridge notes that D is on LAN segment #2 • Bridge forwards frame only on to LAN segment #1

33. DL: Bridges and Spanning Trees Disabled • for increased reliability, desirable to have redundant, alternate paths from source to destination • with multiple simultaneous paths, cycles result - bridges may multiply and forward frame forever • solution: organize bridges in a spanning tree by disabling subset of interfaces

34. DL: Switching • Switches • “multi-port bridge” • Each port acts as a bridge • Each port determines MAC addresses connected to itself • Master list within switch determines forwarding behavior

35. DL: Ethernet Switches • Switching faster: • A-to-B and A’-to-B’ simultaneously, no collisions • layer 2 (frame) filtering using LAN addresses • large number of interfaces versus bridges (which typically have only two) • Flexibly support multiple speeds (10/100/1000) • often: individual hosts, star-connected into switch • Ethernet, but no collisions!

36. DL: Switched Network Advantages • Higher link bandwidth • Point to point electrically simpler than bus • Much greater aggregate bandwidth • Data backplane of switches typically large to support simultaneous transfers amongst port • Can go faster via “cut-through switching” • Frame forwarded from input to output port without awaiting for assembly of entire frame • Slight reduction in latency

37. DL: Bridges vs. Routers • both store-and-forward devices • routers: network layer devices (examine network layer headers) • bridges are Link Layer devices • routers maintain routing tables, implement routing algorithms • bridges maintain filtering tables, implement filtering, learning and spanning tree algorithms • why can't the Internet be one great big bridge?

38. DL: Routers vs. Bridges Bridges + and - + Bridge operation is simpler requiring less processing bandwidth - Topologies are restricted with bridges: a spanning tree must be built to avoid cycles - Bridges do not offer protection from broadcast storms (endless broadcasting by a host will be forwarded by a bridge)

39. DL: Routers vs. Bridges Routers + and - + arbitrary topologies can be supported, cycling is limited by TTL counters (and good routing protocols) + provide protection against broadcast storms - require IP address configuration (not plug and play) - require higher processing bandwidth • bridges do well in small (few hundred hosts) while routers used in large networks (thousands of hosts)

40. Data-link layer summary • principles behind data link layer services: • error detection, correction • sharing a broadcast channel: multiple access • link layer addressing, ARP • various link layer implementations • 802.3 Ethernet • 802.5 Token-ring • 802.11 LANs • PPP • ATM • X.25, Frame Relay • various link layer devices • hubs, bridges, switches

41. Physical Layer • Plethora of physical media • Fiber, copper, air • Specifies the characteristics of transmission media • Too many to cover in detail, not the focus of the course • Many data-link layer protocols (i.e. Ethernet, Token-Ring, FDDI. ATM run across multiple physical layers) • Physical characteristics dictate suitability of data-link layer protocol and bandwidth limits

42. PL: Common Cabling • Copper • Twisted Pair • Unshielded (UTP) • CAT-1, CAT-2, CAT-3, CAT-4, CAT-5, CAT-5e • Shielded (STP) • Coaxial Cable • Fiber • Single-mode • Multi-mode

43. PL: Twisted Pair • Most common LAN interconnection • Multiple pairs of twisted wires • Twisting to eliminate interference • More twisting = Higher data rates, higher cost • Standards specify twisting, resistance, and maximum cable length for use with particular data-link layer

44. PL: Twisted pair • 5 categories • Category 1 • Voice only (telephone wire) • Category 2 • Data to 4Mbs (LocalTalk) • Category 3 • Data to 10Mbs (Ethernet) • Category 4 • Data to 20Mbs (16Mbs Token Ring) • Category 5 (100 MHz) • Data to 100Mbs (Fast Ethernet) • Category 5e (350 MHz) • Data to 1000Mbs (Gigabit Ethernet)

45. PL: Twisted Pair • Common connectors for Twisted Pair • RJ11 (3 pairs) • RJ45 (4 pairs) • Allows both data and phone connections • (1,2) and (3,6) for data, (4,5) for voice • Crossover cables for NIC-NIC, Hub-Hub connection (Data pairs swapped)

46. PL: UTP • Unshielded Twisted Pair • Limited amount of protection from interference • Commonly used for voice and ethernet • Voice: multipair 100-ohm UTP

47. PL: STP • Shielded Twisted Pair • Not as common at UTP • UTP susceptible to radio and electrical interference • Extra shielding material added • Cables heavier, bulkier, and more costly • Often used in token ring topologies • 150 ohm STP two pair (IEEE 802.5 Token Ring)

48. PL: Coaxial cable • Single copper conductor at center • Plastic insulation layer • Highly resistant to interference • Braided metal shield • Support longer connectivity distances over UTP

49. PL: Coaxial cable • Thick (10Base5) • Large diameter 50-ohm cable • N connectors • Thin (10Base2) cables • Small diameter 50-ohm cable • BNC, RJ-58 connector • Video cable • 75-ohm cable • BNC, RJ-59 connector • Not compatible with RJ-58

50. PL: Fiber • Center core made of glass or plastic fiber • Transmit light versus electronic signals • Protects from electronic interference, moisture • Plastic coating to cushion core • Kevlar fiber for strength • Teflon or PVC outer insulating jacket