980 likes | 1.19k Views
GMPLS networks and optical network testbeds. Malathi Veeraraghavan Professor Charles L. Brown Dept. of Electrical & Computer Engineering University of Virginia mvee@virginia.edu Tutorial at ICACT09 Feb. 2009. GMPLS: Generalized MultiProtocol Label Switched networks
E N D
GMPLS networks and optical network testbeds Malathi Veeraraghavan Professor Charles L. Brown Dept. of Electrical & Computer Engineering University of Virginia mvee@virginia.edu Tutorial at ICACT09 Feb. 2009 GMPLS: Generalized MultiProtocol Label Switched networks (MPLS, SONET, WDM, SDM, VLAN)
Outline • Principles • Different types of connection-oriented networks • Technologies • Single network • Internetworking • Usage • Commercial networks • Research & Education Networks (REN)
Principles • Types of switches and networks • Bandwidth sharing modes • TCP in connectionless (IP) networks • Immediate-request and book-ahead modes in connection-oriented networks
Types of networks Connection-oriented
How is bandwidth shared on a connectionless packet-switched network? • Pre-1988 IP network: • Just send data without reservations or any mechanism to adjust rates congestion collapses! • Van Jacobson's 1988 contribution: • Added congestion control to TCP • Sending TCP adjusts rate • Advantages: • Proportional fairness • High utilization • Disadvantages: • No rate guarantees • No temporal fairness (job seniority)
TCP throughput • B: Throughput in congestion-avoidance phase • RTT: Round-trip time • b: an ACK is sent every b segments (b is typically 2) • p: packet loss rate on path • T0: initial retransmission time out in a sequence of retries • Effective rate = min (r,B) • r: bottleneck link rate • Padhye, Firoui, Towsley, Kurose, ACM Sigcomm 98 paper
TCP throughput Case Input parameters Mean transfer delay for a 1GB file (s) Packet loss rate Bottleneck link rate Round-trip delay Case 1 0.0001 100 Mb/s 0.1ms 82.25 Case 2 5ms 89.45 ~21Mbps Case 3 50ms 396.5 Case 4 1Gbps 0.1ms 8.25 Case 5 5ms 39.6 Case 6 50ms 395.7 Case 7 0.001 100 Mbps 0.1ms 82.93 Case 8 5ms 135.4 Case 9 50ms 1293 Case 10 1Gbps 0.1ms 8.64 Case 11 5ms 129.4 Case 12 50ms 1287 Case 13 0.01 100 Mbps 0.1ms 92.41 Case 14 5ms 471.7 ~2Mbps Case 15 50ms 4417 Case 16 1Gbps 0.1ms 12.43 Case 17 5ms 441.7 Case 18 50ms 4387
Bandwidth sharing in circuit networks(immediate-request mode) • Key difference: • Admission control • Intrinsic to circuit networks: position based mux • Send a call setup request: • if requested bandwidth is available, it is allocated to the call • if not, the call is blocked (rejected) • M/G/m/m model: • m: number of circuits
r m ua 4 17 117 24.8% 58.2% 84.6% 1 10 100 ErlangB formula r: offered traffic load in Erlangs : call arrival rate 1/:mean call holding time m: number of circuits Pb: call blocking probability ub: utilization For a 1% call blocking probability, i.e., Pb = 0.01 If m is small, high utilization can only be achieved along with high call blocking probability
Bandwidth sharing mechanismsin CO networks Needed if per-call circuit rate is a large fraction of link capacity (e.g., 1Gbps circuits on a 10Gbps link, m = 10) Bandwidth sharing mechanisms Book-ahead Immediate-request call duration specified unspecified call duration BA-n/BA-First VBDS (Varying-Bandwidth Delayed Start) session-type requests data-type requests BA-n BA-First Users specify a set of call-initiation time options Users are given first available timeslot X. Zhu, Ph.D. Thesis, UVA, http://www.ece.virginia.edu/mv/html-files/students.html
Comparison of Immediate-Request (IR) and Book-Ahead (BA) schemes • Example • To achieve a 90% utilizationwith a call blocking probability less than 10% • BA-First schemes are needed when m < 59 • To achieve a 90% utilization with a call blocking probabilityless than 20% • BA-First schemes are needed when m < 32 U: utilization K: number of time periods in advance-reservation window m=10, K=10, U = 80%: PB = 0.4% BA m=10, U = 80%: PB = 23.6% m=100, U = 80%: PB= 0.4% IR
Virtual circuit (VC) networks Call Admission Control Bandwidth sharing more complex, but better utilization PLUS service guarantees Needed in circuit networks Scheduling (example: weighted fair queueing) Traffic shaping/policing (example: leaky-bucket algorithm) Two additional dimensions in VC networks
Outline • Principles • Different types of connection-oriented networks • Technologies • Single network • Internetworking • Usage • Commercial networks • Research & Education Networks (REN)
Technologies • GMPLS networks • Data-(user-) plane protocols • packet-switched: MPLS, VLAN Ethernet • circuit-switched: SONET/SDH, WDM, SDM (space div. mux) • Control-plane protocols: • RSVP-TE: signaling protocol • OSPF-TE: routing protocol • LMP: link management protocol • Internetworking • GFP, VCAT, LCAS for SONET/SDH • PWE3 for MPLS networks • Digital wrapper for OTN
Label Value CoS S TTL 20 Bits 3 1 8 Multiprotocol label switching (MPLS) • MPLS Header: • Label Value: Label used to identify the virtual circuit • Class of Service (CoS): Experimental field, Used for QoS support • S: Identifies the bottom of the label stack • TTL: Time-To-Live value • Virtual circuits: Label Switched Path (LSP) MPLS Header
IEEE 802.1Q Ethernet VLAN new fields Type/Len Dest. MAC Address Source MAC Address TPID TCI Data FCS FCS: Frame Check Sequence VLAN Tag User Priority 802.1Q Tag Type CFI VLAN ID 2 Bytes 3 Bits 1 Bit 12 Bits
VLAN Tag Fields • Tag Protocol Identifier (TPID) • 802.1Q Tag Protocol Type – set to 0x8100 to identify the frame as a tagged frame • Tag Control Information (TCI) • User Priority • As defined in 802.1p, 3 bits represent eight priority levels • CFI • Canonical Format Indicator, set to indicate the presence of an Embedded-RIF • VLAN ID • Uniquely identifies the frame's VLAN
SONET/SDH rates(number is the multiplier) Example: STS-48 frame has 48 x 90 columns in 125 s STS-1: 90 columns by 9 rows in 125s Tanenbaum
Optical transport networks (OTN) • G. 872 layers • OTS: Optical Transmission Section • OMS: Optical Multiplex Section • OCh: Optical Channel • G.709: • Technique for mapping client signals onto the Optical Channel via layers: • OTU: Optical Channel Transport Unit, and • ODU: Optical Channel Data Unit
Layers within an OTN Courtesy: T. Walker's tutorial
OTN Hierarchy • Electrical domain: • OTU: Optical Channel Transport Unit • ODU: Optical Channel Data Unit • OPU: Optical Channel Payload Unit Low layer Higher layers Courtesy: T. Walker's tutorial
G. 709 Optical Channel frame structure (digital wrapper) • Optical channel (OCh) overhead: support operations, administration, and maintenance functions • OCh payload: can be STM-N, ATM, IP, Ethernet, GFP frames, OTN ODUk, etc. • FEC: Reed-Solomon RS(255, 239) code recommended; roughly introduces a 6.7% overhead • Frame size: 4 rows of 4080 bytes • Frame period: • OTU1 – 48.971 μs (payload data rate: roughly 2.488 Gbps ) • OTU2 – 12.191 μs (payload data rate: roughly 9.995 Gbps ) • OTU3 – 3.035 μs (payload data rate: roughly 40.15 Gbps ) OCh overhead OCh payload FEC
Technologies • GMPLS networks • Data-(user-) plane protocols • packet-switched: MPLS, VLAN Ethernet, Intserv IP • circuit-switched: SONET/SDH, WDM, SDM • Control-plane protocols: • RSVP-TE: signaling protocol • OSPF-TE: routing protocol • LMP: link management protocol • Internetworking • GFP, VCAT, LCAS for SONET/SDH • PWE3 for MPLS networks • Digital wrapper for OTN
The evolution ofResource reSerVation Protocol (RSVP) • RSVP (RFC2205, 1997) • RSVP-TE (RFC 3209, 2001) • RSVP-TE GMPLS Extension (RFC 3471, 3473, 2003) • RSVP-TE GMPLS Extension for SONET/SDH (RFC 3946, 2004, RFC 4606, 2006)
Purpose of signaling(needed only in CO networks) • Functions: • Call setup: • Route selection • Admission control: sufficient bandwidth? • Switch fabric configuration of each switch • recall position based multiplexing • Call release • release bandwidth for use by others
Dest. Next hop III-B III-B III-C III-C Dest. Next hop III-* III Circuit-switched networksPhase 1:Routing protocol exchanges + routing table precomputation • Routing protocols exchange: • topology • address reachability • loading conditions II Host I-A Host III-B I III IV Host III-C Dest. Next hop V III-* IV
a b d c Circuit-switched networksPhase 2: Signaling for call setup Connection setup actions at each switch on the path: • Parse message to extract parameter values • Lookup routing table for next hop to reach destination • Read and update CAC (Connection Admission Control) table • Select timeslots on output port • Configure switch fabric: write entry into timeslot mapping table • Construct setup message to send to next hop Connection setup (Dest: III-B; BW: OC1; Timeslot: a, 1) II b Host I-A a III I Host III-B c b c V IV a d Dest. Next hop Routing table III-* IV
a b d c Circuit-switched networksPhase 2: Signaling for call setup Connection setup actions at each switch on the path: • Parse message to extract parameter values • Lookup routing table for next hop to reach destination • Read and update CAC (Connection Admission Control) table • Select timeslots on output port • Configure switch fabric: write entry into timeslot mapping table • Construct setup message to send to next hop Connection setup (Dest: III-B; BW: OC1; Timeslot: a, 1) II b Host I-A a III I Connection setup Host III-B c b c V IV a d Dest. Next hop Routing table III-* IV Interface (Port); Capacity; Avail timeslots CAC table Next hop c; OC12; 1, 4, 5 IV INPUT Port /Timeslot OUTPUT Port/Timeslot Timeslot mapping table a/1 c/1 Update to remove timeslot 1 from available list
a b d c Circuit-switched networksPhase 2: Signaling for call setup II b Host I-A a Connection setup III I Host III-B c b c V IV a Connection setup (Dest: III-B; BW: OC1; Timeslot: a, 1) d INPUT Port /Timeslot OUTPUT Port/Timeslot Time slot could be different on each hop a/1 c/2 Perform same set of 6 connection setup steps at switch IV write timeslot mapping table entry, update CAC table and send connection setup message to the next hop
a b d c Circuit-switched networksPhase 2: Signaling for call setup INPUT Port /Timeslot OUTPUT Port/Timeslot II d/2 b/1 b Host I-A a Connection setup III I Host III-B c b c V IV a Connection setup d Circuit setup complete Perform same set of 6 connection setup steps at switch III Reverse setup-confirmation messages typically sent from destination through switches to source host
Circuit-switched networksPhase 3: User-data flow • Bits arriving at switch I on time slot 1 at port a are switched to time slot 1 of port c IN Port /Timeslot OUT Port/Timeslot 1 2 II d/2 b/1 b 1 2 1 2 a Host I-A a III I Host III-B b c b d c c 1 2 IV a d V IN Port /Timeslot OUT Port/Timeslot IN Port /Timeslot OUT Port/Timeslot a/1 c/1 a/1 c/2
Release procedure • When a communication session ends, there is a hop-by-hop release procedure (similar to the setup procedure) to release timeslots/wavelengths for use by new calls
RSVP messages and parameters • Messages: • Setup: Path (forward) and Resv (reverse) • Release: PathTear, ResvTear • Parameters • Destination: SESSION object • Bandwidth: Sender Tspec object or SONET/SDH Tspec • Timeslot/Wavelength: • Generalized LABEL for ports, wavelengths • SUKLM label for SONET/SDH • Only supports immediate-request circuits/virtual circuits • No time-dimension parameters for book-ahead
Explicit Route Object (ERO) • A list of groups of nodes along the explicit route (generically called "source route") • Thinking: source routing is better for calls than hop-by-hop routing as it can take into account loading conditions • Constrained shortest path first (CSPF) algorithm executed at the first node to compute end-to-end route, which is included in the ERO
Control-plane message transport: inband or out-of-band • Separation of control plane from data plane in GMPLS networks - out-of-band Internet IP router IP router Control-plane messages Ethernet control ports GMPLS Network Ethernet control ports Circuit established SONET or WDM switch SONET or WDM switch Data-plane link
Interface ID field • Control plane separation: • Requires upstream switch to identify on which data-plane interface the virtual circuit should be routed • Interface ID field defined in the tag-length-value format • Embedded within the RSVP-HOP object • Carried in PATH messages
Technologies • GMPLS networks • Data-(user-) plane protocols • packet-switched: MPLS, VLAN Ethernet, Intserv IP • circuit-switched: SONET/SDH, WDM, SDM • Control-plane protocols: • RSVP-TE: signaling protocol • OSPF-TE: routing protocol • LMP: link management protocol • Internetworking • GFP, VCAT, LCAS for SONET/SDH • PWE3 for MPLS networks • Digital wrapper for OTN
OSPF-TE: Open Shortest Path First -Traffic Engineering • To advertise loading conditions • New parameters: • Maximum bandwidth of a link • Maximum reservable bandwidth: can be greater than the maximum bandwidth to support oversubscription • Unreserved bandwidth • RFC 3630 - for MPLS networks • Only supports immediate-request circuits/virtual circuits • No time-dimension parameters for book-ahead
OSPF-TE extensions for GMPLS • RFC 4202 and 4203 • Main new parameters • Shared Risk Link Group • Interface Switching Capability Descriptor (ISCD) • Allows multiple types of switching techniques • Example for SONET: Minimum LSP Bandwidth: OC1 on a SONET interface if the switch demultiplexes down to OC1 level
Difference between labels in MPLS and circuit-switched GMPLS • In circuit-switched GMPLS networks, labels are not carried in the data plane • Labels in circuit-switched networks identify "position" of data for the circuit - time or wavelength • In circuit-switched GMPLS networks, cannot assign labels without associated bandwidth reservation • In usage section, we will see the value of this feature in MPLS networks • See two applications: traffic engineering, VPLS (addressing benefits)
Technologies • GMPLS networks • Data-(user-) plane protocols • packet-switched: MPLS, VLAN Ethernet, Intserv IP • circuit-switched: SONET/SDH, WDM, SDM • Control-plane protocols: • RSVP-TE: signaling protocol • OSPF-TE: routing protocol • LMP: link management protocol • Internetworking • GFP, VCAT, LCAS for SONET/SDH • PWE3 for MPLS networks • Digital wrapper for OTN
LMP procedures • Control channel management • Set up and maintain control channels between adjacent nodes • Link property correlation • Aggregate multiple data links into a TE link • Synchronize TE link properties at both ends • Link connectivity verification (optional) • Data plane discovery; If_Id exchange; physical connectivity verification • Fault management (optional) • Fault notification and localization Reference: IETF RFC 4204
Control-plane security • Need authentication and integrity for all control-plane exchanges • Since RSVP, OSPF, LMP run over IP, IPsec is a possible solution
Technologies • GMPLS networks • Data-(user-) plane protocols • packet-switched: MPLS, VLAN Ethernet, Intserv IP • circuit-switched: SONET/SDH, WDM, SDM • Control-plane protocols: • RSVP-TE • OSPF-TE • LMP • Internetworking • GFP, VCAT, LCAS for SONET/SDH • PWE3 for MPLS networks • Digital wrapper for OTN
Why internetworking? • GMPLS networks do not exist as standalone entities • Instead they are part of the Internet: • Obvious usage: to interconnect IP routers • Newer uses: • Commercial: interconnect Ethernet switches in geographically distributed LANs via point-to-point links or VPNs • Research & Education networks: connect GbE and 10GbE cards on cluster computers and storage devices to GMPLS networks
Obvious usage • Router-to-router circuits and virtual circuits Internet IP router IP router GMPLS Network SONET or WDM switch SONET or WDM switch
Router-to-router usage • OSPF-enabled usage • simply treat MPLS virtual circuit or GMPLS circuit as a link between routers • allow routing protocol to include these in routing table computations • Data-plane • IP over MPLS • IP over PPP over SONET • Packet-over-SONET (PoS)
Newer uses • New type of gateway functionality • No IP layer involvement • Instead Ethernet frames are mapped onto MPLS virtual circuits or GMPLS circuits • port mapped • VLAN mapped • Cisco and Juniper routers support Ethernet over MPLS • Sycamore and Ciena SONET switches support Ethernet over GMPLS
Ethernet port mapped over MPLS • Send all Ethernet frames received on ports I and II on to the MPLS LSP • MPLS LSP: Pseudo-wire • Enterprise can allocate IP addresses from one subnet: Virtual Private LAN Service (VPLS) • Explains one use for MPLS virtual circuits with no bandwith allocation SDM-to-MPLS gateway SDM-to-MPLS gateway Internet IP router/MPLS switch IP router/MPLS switch Pseudowire II I MPLS LSP (virtual circuit) Ethernet switch Ethernet switch Mux scheme on pseudowire: Ethernet Enterprise 2 Enterprise 1 Gateway: interfaces have different MUX schemes unlike switch, which has same MUX scheme on all links SDM: Space Division Multiplexing