Objective ####################### Scenario to demonstrate conventional IGP Routing --------------------------------------------------------IGP_Routing This scenario demonstrates the conventional IGP used for routing. In this scenario, the nodes are not configured to use Multi Protocol Label Switching (MPLS). Scenario to demonstrate Traffic Engineering (TE) ---------------------------------------------------TE_Improve_Utilizations This scenario demonstrates the use of Traffic Engineering, to better route the traffic along the desired routes. MPLS LSPs are used to specify the routes. As expected, Traffic Engineered network has better throughputs then the one without TE. Scenario using MPLS with Differentiated Services ----------------------------------------------------MPLS_with_Diffserv In this scenario, MPLS LSPs carrying more than one flows use DiffServ codes to assign different QoS levels to different flows. Scenario using Label Distribution Protocol with CSPF ----------------------------------------------------LDP_With_CSPF This scenario demonstrates the use of Label Distribution Protocol (LDP), along with Constrained Shortest Path First (CSPF) for Dynamic Route setups when Dynamic MPLS LSPs are used instead of Static LSPs. Scenario using RSVP-TE with Fast Reroute ----------------------------------------------------RSV_TE_with_Fast_Reroute This Scenario illustrates illustrates the use of RSVP-TE to configure LSPs Dynamicaly, and Use of Fast Reroute to protect LSPs locally. Bypass Tunnel has been configured and Fast Reroute is used to back up LSP at time of link failure in the network. ############################ ######################################### ######################### ##################### ######### Node Models Models Required Required -------------------Wokstation:
Any workstation can be used to generate traffic over the network.
Server: In a client-server architecture, a workstation connects to server for traffic exchange.
Label Edge Router (LER): LER is the edge router used at the ingress and egress point of the MPLS LSPs. Most of the TE and FEC bindings are done at the Ingress LER. The LER is responsible for pushing and popping the SHIM Label onto each packet entering the LSP.
Label Switching Router (LSR): LSRs are the intermediate routers used along the Label Switched Path. The LSRs are responsible for swaping the SHIM Label onto each packet going through the LSP. Note: Any Router in OPNET can be used as LER or LSR. i.e. all routers have MPLS configuration capability.
MPLS Config: This node is used to configure Forwarding Equivalence Class (FEC) and Traffic Trunk specifications. These specifications are associated with different flows at the ingress LERs, to differentiate the flows into various classes and different QoS agreements. --------------------------------------------------------------------------------------------------FEC Specifications: -----------------Used to specify different Forwarding Equivalence Classes to be used in the network. These configurations should be made on the MPLS Config Object. Different FECs can be specified depending upon one or nore combinations of ToS, Protocol used, Source/Destination Address, Source/Destination Port etc.
Traffic Trunk Profiles: ---------------------Traffic Trunk Profiles are specified at the MPLS Config Object. Various Traffic Trunks with different Maximum, average bit rates, and out of profile actions can be specified here. Each Traffic trunk is assigned a DiffServ Class. MPLS Parameters: --------------Used to specify the MPLS parameters to be used. MPLS parameters should be configured at the Label Edge and Label Switching Routers.
Traffic Engineering Configuration: --------------------------------TE Configurations are done at the Ingress Label Edge Routers ( LERs). These configurations are used to perform the traffic bindings. Diferent FEC and Traffic Trunk specifications can be bind to different incoming interfaces at the Router, and they can be assigned to various LSPs.
LDP Parameters: -------------This attribute is used to configure the Label Distribution Protocol (LDP) parameters while using Dynamic LSPs. Neighbour discovery timers, Keep Alive Message timers and Recovery parameters can be specified over here. These configurations should be made on all LERs and LSRs while using Dynamic LSPs. ########################################################################## LSPs Delay Measurement (seconds) -------------------------------LSP Delay (seconds): This statistic can be used to measure End-to-End Delay. It records the delay experienced by Packet in the LSP. i.e. Time spent by the packet within the Label Switched Path.
Flow Delay (seconds): This statistic can be used to measure End-to-End Delay. It records the delay experienced by packet belonging to a specific flow in the LSP. i.e. Time spent by a packet of a given flow inside the Label Switched Path.
Amount of Traffic entering and leaving the LSPs
----------------------------------------------Traffic In (bits/second and packets/second): Total Traffic sent into the LSP at the ingress end of the tunnel. Traffic Out (bits/second and packets/second): Total Traffic received from the LSP at the egress end of the tunnel. Flow Traffic In (bits/second and packets/second): Traffic sent into the LSP at the ingress end of the tunnel for each flow that is carried through this Label Switched Path. Flow Traffic Out (bits/second and packets/second): Traffic received from the LSP at the egress end of the tunnel for each flow that is carried through this Label Switched Path. Utilization of LSPs (Only Dynamic LSPs) ---------------------------------Utilization: Utilization is computed as a percentage of the throughput on the LSP to the amount of Bandwidth allocated when the LSP was setup. Utilization is not be reported if the LSP does not have any constraint to being with and the utilization could be more than 100% if the actual traffic uses more than what was intially reserved.
Flow Utilization: Flow Utilization is computed as a percentage of the throughput of a particular flow on the LSP to the amount of Bandwidth allocated when the LSP was setup. The statistic is reported for each flow that is carried through the Label Switched Path.(i.e., a value indicating how much of the allocated bandwidth is used by any given flow) ############################################################################## FEC: Forwarding Equivalence Class: Class assigned to traffic based on various traffic attributes. This class is used to forward the traffic through different LSPs.
LDP: Label Distribution Protocol: Provides an LSR "discovery" mechanism to enable LSR peers to find each other and establish communication.
CSPF: Constrained Shortest Path First: An algorithm to compute a path for Dynamic LSPs such that path is optimal with respect to some scaler metrics (e.g., minimizes administrative metric, hop count) and does not violate a set of constraints on that LSP
Static LSPs: Label Switched Path configured specifying all the intermediate hops
Dynamic LSPs: Label Switched Path configured specifying only the edge routers or some of the intermediate routers. i.e end-to-end path is loosely defined. ###################################################################### Scenario 1 IGP_Routing NETWORK INFO ============ This network illustrates a typical WAN configuration. A> Background Traffic has been configured from 1> Site 1 2> Site 2 3> Site 3 4> Site 4 5> Site 5 6> Site 6
--> --> --> --> --> -->
B> All edge links are DS3 and Core links are OC3 links. C> No MPLS has been configured in the network. RESULTS ANALYSIS: IGP_Routing
Site 7 and Site 10 Site 7 and Site 11 Site 8 and Site 14 Site 11 and Site 13 Site 7 Site 9
======================================= Throughputs on different links have been shown. Results show that as there is no Traffic Engineering being done, there are certain links with Throughput Zero. ############################################################ 2 NETWORK INFO ============ This Scenario illustrates the use of Traffic Engineering using MPLS to route the traffic in the network to obtain better throughputs. A> Background Traffic has been configured from 1> Site 1 2> Site 2 3> Site 3 4> Site 4 5> Site 5 6> Site 6
--> --> --> --> --> -->
Site 7 and Site 10 Site 7 and Site 11 Site 8 and Site 14 Site 11 and Site 13 Site 7 Site 9
B> All edge links are DS3 and Core links are OC3 links. C> MPLS has been configured in the network. Two FECs have been specified based on the destination address and thus all the traffic going to Site 7 and Site 14 have been routed through diffrent LSPs. RESULTS ANALYSIS: TE_Improve_Utilizations ======================================= Throughputs on different links have been shown. Results show that by doing some Traffic Engineering using MPLS there has been a great change in link Throughputs across the network. Even the links which were not utilized earlier are used, thus increasing overall network Throughputs. ######################################################################### 3 NETWORK INFO ============ This Scenario illustrates the use of MPLS to classify different flows in the LSP, assign them DiffServ Codes and serve them with different QoS Levels. A> Traffic has been configured from 1> Site 3
-->
Site 14
2> Site 4
-->
Site 14
B> All links are DS0 links. C> MPLS has been configured in the network. Two Traffic Trunk Profiles have been specified with different DiffServ codes and the traffic flows from Site 3 and 4 are bind to different trunk profiles at the Ingress LER. D> Routers are configured to perform Weighted Fair Queuing (WFQ).
RESULTS ANALYSIS: MPLS_with_Diffserv ======================================= Upload Response Time, Queuing Delay and Queuing Buffer usage. Results show that even though two similar traffic are flowing to same destination but as different Traffic Trunk Profiles have been assigned to these flows thus the flow with higher DiffServ code has less Upload Response Time, less Queuing Delay and less Queuing Buffer usage. Thus, MPLS can be used to assign different Service Level agreements. ############################################################################## 4 NETWORK INFO ============ This Scenario illustrates the use of LDP and CSPF to configure LSPs Dynamicaly. A> Traffic has been configured from 1> Birmingham 2> Houston --> 3> Los Angeles --> 4> San Jose -->
-->
Seattle San Jose Seattle and New York Tampa
B> All edge links are OC3 and Core links are DS3 links. The ATM core links are OC3. C> MPLS has been configured in the network. Dynamic LSPs are configured by just specifying the Ingress and the Egress Routers. All the constrained path selection is done by LDP and CSPF. D> Routers are configured to use LDP for neighbour discovery. E> Constraints have been set up on some LSPs in order to illustrate that the normal IGP routes need not be used by CSPF if constraints are not met.
- Birmingham to Seattle: The color Blue is excluded in its resource class configuration. This color is supported by the interfaces connecting Atlanta to Dallas and Oklahoma City. Therefore, this LSP takes a longer route. - Houston to San Jose: This LSP is configured to have a bandwidth requirement of 1 Mbps. The sub-interface connecting Dallas to Salt Lake City is on the shortest path, but it is configured with a bandwidth metric of 500 Kbps, which is insufficient for the LSP. Hence the LSP is routed via Denver. The sub-interfaces that are used to route the LSP have a default bandwidth metric, which is equal to the links connected to the parent interfaces.
RESULTS ANALYSIS: LDP_with_CSPF ======================================= Traffic In and Out of the LSPs in the network have been shown. With the use of LDP and CSPF, just Ingress and Egress Routers are specified and all the constrained path selection is performed automatically. In cases where the shortest path does not satisfy the configured constraints, the best suitable path is selected. The LSP from Birmigham to Seattle is routed via Washington DC since the links from Atlanta to Dallas and Oklahoma City support a resource class that is excluded on the LSP. The LSP from Houston to San Jose is routed on the sub-interface connecting Dallas to Denver and then on to Salt Lake City, since the sub-interface connecting Dallas to Salt Lake City has a configured bandwidth of 500 Kbps, which is too less for the LSP. ############################################################### 5 NETWORK INFO ============ This Scenario illustrates the use of RSVP-TE to configure LSPs Dynamicaly. And Use of Fast Reroute to protect LSPs locally.
A> Traffic has been configured from 1> San Francisco 2> Los Angeles
--> -->
Maine Maine
B> All links are OC3 links. C> MPLS has been configured in the network. Two Primary Dynamic LSPs are configured from Las Vegas to Boston via Houston. D> One Primary LSP use Ingress initiated Backup LSP for protection and other primary LSP use a Bypass Tunnel to protect locally.
E> Routers are configured to use RSVP-TE for setting up LSPs. F> Link between "Cincinnati to Washington" fails at time 450 seconds.
RESULTS ANALYSIS: RSVP-TE with Fast Reroute ======================================= 1> Traffic reroute time has been shown for the primary LSPs. Results show that for LSP that is using Fast Reroute for protection with Bypass Tunnels, Traffic reroute time is almost 10 times less then the Traffic reroute time for LSP that use Ingress initiated Backup LSP.
2> Traffic In of the LSPs in the network have also been shown. Results show that Traffic is immediately bypassed using Bypass Tunnel for the LSP that uses Fast Reroute. Whereas, Traffic is switched to backup LSP some time after failure, for LSP that use Ingress initiated Backup LSP. ############################################################################## VPN
PROJECT DESCRIPTION =================== This project illustrates some of the various technologies that can be used to set up Virtual Private Networks (VPNs). A brief description of the scenarios is given here. Note: Each scenario contains a detailed description of that particular setup.
VPNs using IP Tunnels ===================== The scenario "VPNs_with_IP_Tunnels" illustrates the use of layer-3 tunnels (GRE/IP-IP) to connect remote sites to a headquarters site. IGP routing protocols like EIGRP, OSPF etc can be run on the tunnels to exchange routing information between the various VPN sites. MPLS/BGP VPNs ============= These VPNs are based on RFC-2547. MPLS is used for forwarding packets over the backbone, and BGP is used for distributing routes over the backbone. This project contains two scenarios that illustrate MPLS/BGP VPNs. In the scenario "MPLS_VPNs_with_Static_Routes", static routes are used to announce the
customer edge (CE) routes into the VRFs. In the scenario "MPLS_VPNs_with_BGP", EBGP is used as the PE-CE (provider edge - customer edge) protocol. Note: To return to the top level, right click on the blank screen and select "Go to Parent Subnet" #################################################################### MPLS_VPN_with_SR Network Configuration NETWORK INFO ============ This Scenario illustrates the use VPNs for communication between several different sites. Two different Enterprise networks have been shown: Enterprise A and Enterprise B. Enterprise A uses a VPN named "YELLOW_VPN" and Enterprise B uses a VPN named "RED_VPN" to communicate between different sits. A> Background Traffic has been configured from 1> Enterprise A: Site 1: A_1_Rtr1 --> Enterprise A: Site 2: A_2_Rtr1 2> Enterprise B: Site 1: B_1_Rtr1 --> Enterprise B: Site 2: B_2_Rtr1 B> All links are PPP_SONET_OC3 links. C> LSPs has been configured between each PE in the network. D> BGP is configured between all PEs, i.e. all PEs are BGP neighbors. E> Routes between PE and CE have been statically configured. F> All sites in Enterprise A are configured to use "YELLOW_VPN" and all sites in Enterprise B are configured to use "RED_VPN".
RESULTS ANALYSIS: VRF Tables ============================ VRF tables created by Site1_PE are shown here. (VRF tables can be exported to Simulation Log) Comparing the two VRF tables show that: 1> Enterprise A and Enterprise B are using same address spaces and have similar IP addresses across the network.
2> As, VPNs are being used, the labels (Bottom Label) being assigned to entries in the VRF tables are different. This isolates traffic from YELLOW_VPN and RED_VPN. 3> The VRF Tables show, a> Destination Address b> Next BGP Hop address, i.e. next PE. c> Bottom Labels, to differentiate the VPNs d> Top Labels to be used to move traffic across the LSPs.
===================================================================== ================================= VRF TABLE snapshot for: Router name: Logical Network.Site1_PE at time: 3600.00 seconds VRF TABLE contents:
VRF Table for VPN at Logical Network.Site1_PE: Destination
Subnet Mask
Next BGP Hop
Out Iface
Bottom Label
Top
Label ----------192.0.31.0 192.0.33.0 192.0.32.0 192.0.28.0 192.0.29.0 192.0.30.0 192.0.34.0 192.0.35.0 192.0.36.0
------------------------------ -----------255.255.255.0 192.0.17.2 IF0 31 255.255.255.0 192.0.17.2 IF0 32 255.255.255.0 192.0.17.2 IF0 33 255.255.255.0 192.0.14.1 IF2 31 255.255.255.0 192.0.14.1 IF2 32 255.255.255.0 192.0.14.1 IF2 33 255.255.255.0 192.0.16.2 IF2 31 255.255.255.0 192.0.16.2 IF2 32 255.255.255.0 192.0.16.2 IF2 33
--------UNDEF UNDEF UNDEF 34 34 34 31 31 31
VRF Table for VPN at Logical Network.Site1_PE: Destination
Subnet Mask
Next BGP Hop
Out Iface
Bottom Label
Label ----------192.0.31.0 192.0.32.0 192.0.33.0 192.0.28.0
------------------------------ ------------ --------255.255.255.0 192.0.100.11 IF3 34 UNDEF 255.255.255.0 192.0.100.11 IF3 35 UNDEF 255.255.255.0 192.0.100.11 IF3 36 UNDEF 255.255.255.0 192.0.14.1 IF2 34 34
Top
192.0.29.0 192.0.30.0 192.0.34.0 192.0.35.0 192.0.36.0
255.255.255.0 255.255.255.0 255.255.255.0 255.255.255.0 255.255.255.0
192.0.14.1 192.0.14.1 192.0.16.2 192.0.16.2 192.0.16.2
IF2 IF2 IF2 IF2 IF2
35 36 34 35 36
34 34 31 31 31
####################################################################### MPLS_VPN_with_BGP NETWORK INFO ============ This Scenario illustrates the use VPNs for communication between several different sites. Two different Eneterprise networks have been shown: Enterprise A and Enterprise B. Enterprise A uses a VPN named "YELLOW_VPN" and Enterprise B uses a VPN named "RED_VPN" to communicate between different sits. A> Background Traffic has been configured from 1> Enterprise A: Site 1: A_1_Rtr1 --> Enterprise A: Site 2: A_2_Rtr1 2> Enterprise B: Site 1: B_1_Rtr1 --> Enterprise B: Site 2: B_2_Rtr1 B> All links are PPP_SONET_OC3 links. C> LSPs has been configured between each PE in the network. D> BGP is configured between all PEs, i.e. all PEs are BGP neighbors. E> Routes between PE and CE are shared using BGP, i.e. PEs and CEs are BGP neighbors of each other. F> All sites in Enterprise A are configured to use "YELLOW_VPN" and all sites in Enterprise B are configured to use "RED_VPN".
RESULTS ANALYSIS: VRF Tables ============================ VRF tables created by Site1_PE are shown here. (VRF tables can be exported to Simulation Log) Comparing the two VRF tables show that: 1> Enterprise A and Enterprise B are using same address spaces and have similar IP addresses across the network.
2> As, VPNs are being used, the labels (Bottom Label) being assigned to entries in the VRF tables are different. This isolates traffic from YELLOW_VPN and RED_VPN. 3> The VRF Tables show, a> Destination Address b> Next BGP Hop address, i.e. next PE. c> Bottom Labels, to differentiate the VPNs d> Top Labels to be used to move traffic across the LSPs.
===================================================================== ================================= VRF TABLE snapshot for: Router name: Logical Network.Site1_PE at time: 3600.00 seconds VRF TABLE contents:
VRF Table for VPN at Logical Network.Site1_PE: Destination
Subnet Mask
Next BGP Hop
Out Iface
Bottom Label
Label ----------192.0.4.0 192.0.2.0 192.0.31.0 192.0.3.0 192.0.33.0 192.0.32.0 192.0.7.0 192.0.6.0 192.0.28.0 192.0.5.0 192.0.30.0 192.0.29.0 192.0.21.0 192.0.19.0 192.0.34.0 192.0.20.0 192.0.36.0 192.0.35.0
------------------------------ -----------255.255.255.0 192.0.17.2 IF0 37 255.255.255.0 192.0.17.2 IF0 38 255.255.255.0 192.0.17.2 IF0 39 255.255.255.0 192.0.17.2 IF0 40 255.255.255.0 192.0.17.2 IF0 41 255.255.255.0 192.0.17.2 IF0 42 255.255.255.0 192.0.25.1 IF2 31 255.255.255.0 192.0.25.1 IF2 32 255.255.255.0 192.0.25.1 IF2 33 255.255.255.0 192.0.25.1 IF2 34 255.255.255.0 192.0.25.1 IF2 35 255.255.255.0 192.0.25.1 IF2 36 255.255.255.0 192.0.26.1 IF2 33 255.255.255.0 192.0.26.1 IF2 34 255.255.255.0 192.0.26.1 IF2 35 255.255.255.0 192.0.26.1 IF2 36 255.255.255.0 192.0.26.1 IF2 37 255.255.255.0 192.0.26.1 IF2 38
VRF Table for VPN at Logical Network.Site1_PE:
--------UNDEF UNDEF UNDEF UNDEF UNDEF UNDEF 34 34 34 34 34 34 31 31 31 31 31 31
Top
Destination
Subnet Mask
Next BGP Hop
Out Iface
Bottom Label
Top
Label ---------------------------------------- ------------ --------192.0.4.0 255.255.255.0 192.0.100.11 IF3 31 UNDEF 192.0.2.0 255.255.255.0 192.0.100.11 IF3 32 UNDEF 192.0.231.0 255.255.255.0 192.0.100.11 IF3 33 UNDEF 192.0.3.0 255.255.255.0 192.0.100.11 IF3 34 UNDEF 192.0.33.0 255.255.255.0 192.0.100.11 IF3 35 UNDEF 192.0.32.0 255.255.255.0 192.0.100.11 IF3 36 UNDEF 192.0.7.0 255.255.255.0 192.0.25.1 IF2 37 34 192.0.6.0 255.255.255.0 192.0.25.1 IF2 38 34 192.0.228.0 255.255.255.0 192.0.25.1 IF2 39 34 192.0.5.0 255.255.255.0 192.0.25.1 IF2 40 34 192.0.30.0 255.255.255.0 192.0.25.1 IF2 41 34 192.0.29.0 255.255.255.0 192.0.25.1 IF2 42 34 192.0.21.0 255.255.255.0 192.0.26.1 IF2 39 31 192.0.19.0 255.255.255.0 192.0.26.1 IF2 40 31 192.0.234.0 255.255.255.0 192.0.26.1 IF2 41 31 192.0.20.0 255.255.255.0 192.0.26.1 IF2 42 31 192.0.36.0 255.255.255.0 192.0.26.1 IF2 43 31 192.0.35.0 255.255.255.0 192.0.26.1 IF2 44 31
##########################################################################
