1 / 120

Network Layer and Data Center Topologies

Network Layer and Data Center Topologies. Hakim Weatherspoon Assistant Professor, Dept of Computer Science CS 5413: High Performance Systems and Networking September 8, 2014. Slides used and adapted judiciously from Computer Networking, A Top-Down Approach. Goals for Today.

brianmorris
Télécharger la présentation

Network Layer and Data Center Topologies

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Network Layer and Data Center Topologies Hakim Weatherspoon Assistant Professor, Dept of Computer Science CS 5413: High Performance Systems and Networking September 8, 2014 Slides used and adapted judiciously from Computer Networking, A Top-Down Approach

  2. Goals for Today • Network Layer • Abstraction / services • Datagram vs Virtual Circuit (VC) • Internet Protocol • IP Datagram format • IP Addressing • Hierarchical Routing • Data Center Topologies • FatTree • Backup Slides • DHCP and NAT • ICMP and Traceroute • IPv6 • Hierarchical Routing: RIP, OSPF, BGP

  3. transport segment from sending to receiving host on sending side encapsulates segments into datagrams on receiving side, delivers segments to transport layer network layer protocols in everyhost, router router examines header fields in all IP datagrams passing through it Network Layer network data link physical network data link physical network data link physical network data link physical network data link physical network data link physical network data link physical network data link physical network data link physical network data link physical network data link physical application transport network data link physical application transport network data link physical

  4. Network Layer Two key functions • forwarding: move packets from router’s input to appropriate router output • routing: determine route taken by packets from source to dest. • routing algorithms analogy: • routing: process of planning trip from source to dest • forwarding: process of getting through single interchange

  5. Network Layer routing algorithm local forwarding table header value output link routing algorithm determines end-end-path through network forwarding table determines local forwarding at this router 0100 0101 0111 1001 3 2 2 1 value in arriving packet’s header 1 0111 2 3 Interplay between routing and forwarding

  6. Goals for Today • Network Layer • Abstraction / services • Datagram vs Virtual Circuit (VC) • Internet Protocol • IP Datagram format • Addressing • Hierarchical Routing • Data Center Topologies • FatTree • Backup Slides • DHCP and NAT • ICMP and Traceroute • IPv6 • Hierarchical Routing: RIP, OSPF, BGP

  7. Network Layer Connection, Connection-less services • datagram network provides network-layer connectionless service • virtual-circuit network provides network-layer connection service • analogous to TCP/UDP connecton-oriented / connectionless transport-layer services, but: • service:host-to-host • no choice:network provides one or the other • implementation:in network core

  8. call setup, teardown for each call before data can flow each packet carries VC identifier (not destination host address) every router on source-dest path maintains “state” for each passing connection link, router resources (bandwidth, buffers) may be allocated to VC (dedicated resources = predictable service) “source-to-dest path behaves much like telephone circuit” performance-wise network actions along source-to-dest path Network Layer Virtual Circuits (VC)

  9. Network Layer Virtual Circuits (VC) implementation a VC consists of: • path from source to destination • VC numbers, one number for each link along path • entries in forwarding tables in routers along path • packet belonging to VC carries VC number (rather than dest address) • VC number can be changed on each link. • new VC number comes from forwarding table

  10. Network Layer Virtual Circuits (VC) forwarding table 22 32 12 3 1 2 VC number interface number forwarding table in northwest router: Incoming interface Incoming VC # Outgoing interface Outgoing VC # 1 12 3 22 2 63 1 18 3 7 2 17 1 97 3 87 … … … … VC routers maintain connection state information!

  11. used to setup, maintain teardown VC used in ATM, frame-relay, X.25 not used in today’s Internet Network Layer application transport network data link physical application transport network data link physical Virtual Circuits (VC) signaling protocol 6. receive data 5. data flow begins 4. call connected 3. accept call 1. initiate call 2. incoming call

  12. no call setup at network layer routers: no state about end-to-end connections no network-level concept of “connection” packets forwarded using destination host address Network Layer application transport network data link physical application transport network data link physical Datagram Networks 1. send datagrams 2. receive datagrams

  13. Network Layer 4 billion IP addresses, so rather than list individual destination address list range of addresses (aggregate table entries) 1 3 2 Datagram Forwarding Table routing algorithm local forwarding table dest address output link address-range 1 address-range 2 address-range 3 address-range 4 3 2 2 1 IP destination address in arriving packet’s header

  14. Network Layer Datagram Forwarding Table Destination Address Range 11001000 00010111 00010000 00000000 through 11001000 00010111 00010111 11111111 11001000 00010111 00011000 00000000 through 11001000 00010111 00011000 11111111 11001000 00010111 00011001 00000000 through 11001000 00010111 00011111 11111111 otherwise Link Interface 0 1 2 3 Q: but what happens if ranges don’t divide up so nicely?

  15. Network Layer Datagram Forwarding Table: Longest Prefix Matching longest prefix matching when looking for forwarding table entry for given destination address, use longest address prefix that matches destination address. Link interface 0 1 2 3 Destination Address Range 11001000 00010111 00010*** ********* 11001000 00010111 00011000 ********* 11001000 00010111 00011*** ********* otherwise examples: which interface? DA: 11001000 00010111 00010110 10100001 which interface? DA: 11001000 00010111 00011000 10101010

  16. Internet (datagram) data exchange among computers “elastic” service, no strict timing req. many link types different characteristics uniform service difficult “smart” end systems (computers) can adapt, perform control, error recovery simple inside network, complexity at “edge” ATM (VC) evolved from telephony human conversation: strict timing, reliability requirements need for guaranteed service “dumb” end systems telephones complexity inside network Network Layer Datagram versus Virtual Circuits (VC)

  17. Goals for Today • Network Layer • Abstraction / services • Datagram vs Virtual Circuit (VC) • Internet Protocol • IP Datagram format • IP Addressing • Hierarchical Routing • Data Center Topologies • FatTree • Backup Slides • DHCP and NAT • ICMP and Traceroute • IPv6 • Hierarchical Routing: RIP, OSPF, BGP

  18. host, router network layer functions: The Internet Protocol Network Layer • IP protocol • addressing conventions • datagram format • packet handling conventions forwarding table transport layer: TCP, UDP • routing protocols • path selection • RIP, OSPF, BGP network layer • ICMP protocol • error reporting • router “signaling” link layer physical layer

  19. The Internet Protocol Network Layer IP protocol version number 32 bits total datagram length (bytes) header length (bytes) type of service head. len ver length for fragmentation/ reassembly fragment offset “type” of data flgs 16-bit identifier max number remaining hops (decremented at each router) upper layer time to live header checksum 32 bit source IP address 32 bit destination IP address upper layer protocol to deliver payload to e.g. timestamp, record route taken, specify list of routers to visit. options (if any) data (variable length, typically a TCP or UDP segment) IP Datagram format how much overhead? • 20 bytes of TCP • 20 bytes of IP • = 40 bytes + app layer overhead

  20. network links have MTU (max.transfer size) - largest possible link-level frame different link types, different MTUs large IP datagram divided (“fragmented”) within net one datagram becomes several datagrams “reassembled” only at final destination IP header bits used to identify, order related fragments The Internet Protocol Network Layer … … reassembly IP Fragmentation/Reassembly fragmentation: in: one large datagram out: 3 smaller datagrams

  21. The Internet Protocol Network Layer length =1040 length =4000 length =1500 length =1500 ID =x ID =x ID =x ID =x fragflag =0 fragflag =1 fragflag =1 fragflag =0 offset =0 offset =370 offset =0 offset =185 one large datagram becomes several smaller datagrams IP Fragmentation/Reassembly example: • 4000 byte datagram • MTU = 1500 bytes 1480 bytes in data field offset = 1480/8

  22. Goals for Today • Network Layer • Abstraction / services • Datagram vs Virtual Circuit (VC) • Internet Protocol • IP Datagram format • IP Addressing • Hierarchical Routing • Data Center Topologies • FatTree • Backup Slides • DHCP and NAT • ICMP and Traceroute • IPv6 • Hierarchical Routing: RIP, OSPF, BGP

  23. IP address: 32-bit identifier for host, router interface interface: connection between host/router and physical link router’s typically have multiple interfaces host typically has one or two interfaces (e.g., wired Ethernet, wireless 802.11) IP addresses associated with each interface The Internet Protocol Network Layer 223.1.1.2 223.1.3.27 IP Addressing 223.1.1.1 223.1.2.1 223.1.1.4 223.1.2.9 223.1.1.3 223.1.2.2 223.1.3.2 223.1.3.1 223.1.1.1 = 11011111 00000001 00000001 00000001 223 1 1 1

  24. Q: how are interfaces actually connected? A: we’ll learn about that in chapter 5, 6. The Internet Protocol Network Layer 223.1.1.2 223.1.3.27 IP Addressing 223.1.1.1 223.1.2.1 223.1.1.4 223.1.2.9 223.1.1.3 223.1.2.2 A: wired Ethernet interfaces connected by Ethernet switches 223.1.3.2 223.1.3.1 For now: don’t need to worry about how one interface is connected to another (with no intervening router) A: wireless WiFi interfaces connected by WiFi base station

  25. IP address: subnet part - high order bits host part - low order bits what’s a subnet ? device interfaces with same subnet part of IP address can physically reach each other without intervening router The Internet Protocol Network Layer subnet Subnets 223.1.1.1 223.1.2.1 223.1.1.2 223.1.1.4 223.1.2.9 223.1.2.2 223.1.3.27 223.1.1.3 223.1.3.2 223.1.3.1 network consisting of 3 subnets

  26. recipe to determine the subnets, detach each interface from its host or router, creating islands of isolated networks each isolated network is called a subnet The Internet Protocol Network Layer 223.1.1.0/24 223.1.2.0/24 223.1.1.1 223.1.2.1 223.1.1.2 223.1.1.4 223.1.2.9 223.1.2.2 223.1.3.27 223.1.1.3 223.1.3.2 223.1.3.1 223.1.3.0/24 subnet Subnets subnet mask: /24

  27. how many? The Internet Protocol Network Layer 223.1.1.2 Subnets 223.1.1.1 223.1.1.4 223.1.1.3 223.1.7.0 223.1.9.2 223.1.9.1 223.1.7.1 223.1.8.1 223.1.8.0 223.1.2.6 223.1.3.27 223.1.2.1 223.1.2.2 223.1.3.1 223.1.3.2

  28. The Internet Protocol Network Layer IP Addressing: CIDR (ClassessInterDomain Routing) CIDR:Classless InterDomain Routing • subnet portion of address of arbitrary length • address format: a.b.c.d/x, where x is # bits in subnet portion of address host part subnet part 11001000 00010111 00010000 00000000 200.23.16.0/23

  29. The Internet Protocol Network Layer IP Addresses: How to get one? Q: How does a host get IP address? • hard-coded by system admin in a file • Windows: control-panel->network->configuration->tcp/ip->properties • UNIX: /etc/rc.config • DHCP:Dynamic Host Configuration Protocol: dynamically get address from as server • “plug-and-play”

  30. Goals for Today • Network Layer • Abstraction / services • Datagram vs Virtual Circuit (VC) • Internet Protocol • IP Datagram format • IP Addressing / subnets • Routing Algorithms • Hierarchical Routing • Data Center Topologies • FatTree • Backup Slides • DHCP and NAT • ICMP and Traceroute • IPv6 • Hierarchical Routing: RIP, OSPF, BGP

  31. The Internet Protocol Network Layer routing algorithm determines end-end-path through network forwarding table determines local forwarding at this router 1 3 2 Interplay between routing and forwarding routing algorithm local forwarding table dest address output link address-range 1 address-range 2 address-range 3 address-range 4 3 2 2 1 IP destination address in arriving packet’s header

  32. The Internet Protocol Network Layer 5 3 5 2 2 1 3 1 2 1 z x w u y v Graph Abstractions graph: G = (N,E) N = set of routers = { u, v, w, x, y, z } E = set of links ={ (u,v), (u,x), (v,x), (v,w), (x,w), (x,y), (w,y), (w,z), (y,z) } aside: graph abstraction is useful in other network contexts, e.g., P2P, where N is set of peers and E is set of TCP connections

  33. The Internet Protocol Network Layer 5 3 5 2 2 1 3 1 2 1 z x w u y v Graph Abstractions: Costs c(x,x’) = cost of link (x,x’) e.g., c(w,z) = 5 cost could always be 1, or inversely related to bandwidth, or inversely related to congestion cost of path (x1, x2, x3,…, xp) = c(x1,x2) + c(x2,x3) + … + c(xp-1,xp) key question: what is the least-cost path between u and z ? routing algorithm: algorithm that finds that least cost path

  34. Q: global or decentralized information? global: all routers have complete topology, link cost info “link state” algorithms decentralized: router knows physically-connected neighbors, link costs to neighbors iterative process of computation, exchange of info with neighbors “distance vector” algorithms Q: static or dynamic? static: routes change slowly over time dynamic: routes change more quickly periodic update in response to link cost changes The Internet Protocol Network Layer Routing Algorithm Classifications

  35. Goals for Today • Network Layer • Abstraction / services • Datagram vs Virtual Circuit (VC) • Internet Protocol • IP Datagram format • IP Addressing / subnets • Routing Algorithms • Hierarchical Routing • Data Center Topologies • FatTree • Backup Slides • DHCP and NAT • ICMP and Traceroute • IPv6 • Hierarchical Routing: RIP, OSPF, BGP

  36. scale: with 600 million destinations: can’t store all dest’s in routing tables! routing table exchange would swamp links! administrative autonomy internet = network of networks each network admin may want to control routing in its own network The Internet Protocol Network Layer Hierarchical Routing our routing study thus far - idealization • all routers identical • network “flat” … not true in practice

  37. aggregate routers into regions,“autonomous systems” (AS) routers in same AS run same routing protocol “intra-AS” routing protocol routers in different AS can run different intra-AS routing protocol gateway router: at “edge” of its own AS has link to router in another AS The Internet Protocol Network Layer Hierarchical Routing

  38. forwarding table configured by both intra- and inter-AS routing algorithm intra-AS sets entries for internal dests inter-AS & intra-AS sets entries for external dests The Internet Protocol Network Layer 3a 3b 2a AS3 AS2 1a 2c AS1 2b 1b 1d 1c 3c Inter-AS Routing algorithm Intra-AS Routing algorithm Forwarding table Hierarchical Routing: Interconnected Autonomous Systems (AS)

  39. The Internet Protocol Network Layer Hierarchical Routing: Intra-AS routing • also known as interior gateway protocols (IGP) • most common intra-AS routing protocols: • RIP: Routing Information Protocol • OSPF: Open Shortest Path First • IGRP: Interior Gateway Routing Protocol (Cisco proprietary)

  40. The Internet Protocol Network Layer Hierarchical Routing: Inter-AS routing—BGP • BGP (Border Gateway Protocol):the de facto inter-domain routing protocol • “glue that holds the Internet together” • BGP provides each AS a means to: • eBGP: obtain subnet reachability information from neighboring ASs. • iBGP: propagate reachability information to all AS-internal routers. • determine “good” routes to other networks based on reachability information and policy. • allows subnet to advertise its existence to rest of Internet: “I am here”

  41. The Internet Protocol Network Layer Hierarchical Routing: Inter-AS routing—BGP Path Attributes and BGP Routes • advertised prefix includes BGP attributes • prefix + attributes = “route” • two important attributes: • AS-PATH: contains ASs through which prefix advertisement has passed: e.g., AS 67, AS 17 • NEXT-HOP: indicates specific internal-AS router to next-hop AS. (may be multiple links from current AS to next-hop-AS) • gateway router receiving route advertisement uses import policy to accept/decline • e.g., never route through AS x • policy-basedrouting

  42. The Internet Protocol Network Layer Hierarchical Routing: Inter-AS routing—BGP BGP Route Selection • router may learn about more than 1 route to destination AS, selects route based on: • local preference value attribute: policy decision • shortest AS-PATH • closest NEXT-HOP router: hot potato routing • additional criteria

  43. The Internet Protocol Network Layer Hierarchical Routing: Inter-AS routing—BGP BGP Messages • BGP messages exchanged between peers over TCP connection • BGP messages: • OPEN: opens TCP connection to peer and authenticates sender • UPDATE:advertises new path (or withdraws old) • KEEPALIVE: keeps connection alive in absence of UPDATES; also ACKs OPEN request • NOTIFICATION: reports errors in previous msg; also used to close connection

  44. The Internet Protocol Network Layer legend: provider B network X W A customer network: C Y Hierarchical Routing: Inter-AS routing—BGP BGP Routing Policy • A,B,C are provider networks • X,W,Y are customer (of provider networks) • X is dual-homed: attached to two networks • X does not want to route from B via X to C • .. so X will not advertise to B a route to C

  45. The Internet Protocol Network Layer legend: provider B network X W A customer network: C Y Hierarchical Routing: Inter-AS routing—BGP BGP Routing Policy • A advertises path AW to B • B advertises path BAW to X • Should B advertise path BAW to C? • No way! B gets no “revenue” for routing CBAW since neither W nor C are B’s customers • B wants to force C to route to w via A • B wants to route onlyto/from its customers!

  46. The Internet Protocol Network Layer Hierarchical Routing: Intra vs Inter-AS routing policy: • inter-AS: admin wants control over how its traffic routed, who routes through its net. • intra-AS: single admin, so no policy decisions needed scale: • hierarchical routing saves table size, reduced update traffic performance: • intra-AS: can focus on performance • inter-AS: policy may dominate over performance

  47. Goals for Today • Network Layer • Abstraction / services • Datagram vs Virtual Circuit (VC) • Internet Protocol • IP Datagram format • IP Addressing / subnets • Routing Algorithms • Hierarchical Routing • Data Center Topologies • FatTree • Backup Slides • DHCP and NAT • ICMP and Traceroute • IPv6 • Hierarchical Routing: RIP, OSPF, BGP

  48. Data Center Topology: FatTree • A scalable, commodity data center network architecture, M. Al-Fares, A. Loukissas, and A. Vahdat. ACM SIGCOMM Computer Communication Review, Volume 38, Issue 4 (August 2008), pages 63-74.

  49. Data Center Topology: FatTree Overview • Structure and Properties of a Data Center • Desired properties in a DC Architecture • Fat tree based solution

  50. Data Center Topology: FatTree Background • Topology: • 2 layers: 5K to 8K hosts • 3 layer: >25K hosts • Switches: • Leaves: have N GigE ports (48-288) + N 10 GigE uplinks to one or more layers of network elements • Higher levels: N 10 GigE ports (32-128) • Multi-path Routing: • Ex. ECMP • without it, the largest cluster = 1,280 nodes • Performs static load splitting among flows • Lead to oversubscription for simple comm. patterns • Routing table entries grows multiplicatively with number of paths, cost ++, lookup latency ++

More Related