Library: Test Plans

MPLS Testing

1. IxANVL LDP Conformance Test
2. IxScriptMate L2 VPN Partial Mesh Performance Test
3. IxScriptMate L3 VPN RSVP-TE Egress Performance
4. IxExplorer LDP Extended Martini Session Scalability Test
5. IxScriptMate VPLS MAC Address Cache Capacity Test
6. IxExplorer RSVP-TE Fast Reroute Test

1. IxANVL LDP Conformance Test

Objective: To characterize the MPLS router's compliance with LDP RFC standards.

Test setup: IxANVL Linux workstation connected directly, or via Ixia test hardware, to DUT with two network connections one for request packets and one for response packets.

Methodology: IxANVL runs a number of test cases against the DUT based on direct interpretation of the LDP RFCs 3036 and 3215.

1. Configure each IxANVL network interface with the appropriate network parameters, including those of the DUT such as IP address, MAC address, gateway, etc.
2. Specify configuration of the DUT, typically via command scripts such as Expect scripts.
3. Select the set of IxANVL LDP test cases to run during the test session (see Figure 1).
4. Run IxANVL in batch mode with the command scripts, reconfiguring the DUT as required between test cases to match the IxANVL test setup.

Results: Number of tests passed/failed, including reasons for failed cases (see Figure 2).


Figure 1. IxANVL LDP test cases.


Figure 2. IxANVL LDP test results.

2. IxScriptMate L2 VPN Partial Mesh

Performance Test

Objective: To determine the maximum rate an LSR configured as an L2 VPN PE router can push MPLS labels onto incoming Layer 2 packets and forward the resulting MPLS traffic, or pop labels from incoming MPLS packets and forward the resulting Layer 2 traffic with no loss.

Test setup: Minimum of two Ixia ports connected to the DUT one to simulate a PE router and another to simulate a CE router. Additional ports can be configured to emulate multiple PEs/CEs (see Figure 3). The IxScriptMate application, running on a workstation, controls the test running on Ixia hardware.

Parameters: Number of PEs/CEs, number of VCs per PE, VC distribution style (consecutive or round robin), frame size distribution, traffic transmission rate.

Methodology:

1. Each Ixia port simulating a PE router establishes an OSPF session, LDP session, and Extended LDP (Martini) session with the DUT.
2. If more than one Ixia port is being used to simulate PE routers, the VC IDs can be distributed among them in either round robin or consecutive fashion.
3. The CE and/or PE ports transmit traffic at the specified rate. If frame loss occurs, the transmission rate is alternately reduced/increased using a binary search algorithm to determine the maximum rate at which the DUT can forward traffic without loss

Results: Frame loss, latency, data errors, sequence errors.


Figure 3. Test setup

3. IxScriptMate L3 VPN RSVP-TE Egress Performance

Objective: To determine the maximum rate an LER configured as an L3 VPN PE router can pop MPLS labels from incoming packets and forward the resulting IP traffic with no loss.

Test setup: Two Ixia ports connected to the DUT one to simulate a PE and P router and another to simulate a CE router (see Figure 4).

Parameters: OSPF parameters, BGP parameters, number of routes per CE, frame size distribution, traffic transmission rate.

Methodology:

1. The port simulating the PE/P router establishes an OSPF session with the DUT and advertises the loopback address of the simulated PE router. Traffic engineering parameters are advertised using OSPF-TE.
2. Bidirectional LSPs are established using RSVP-TE.
3. A Multi-Protocol internal BGP (MP iBGP) session is established with the DUT.
4. The port simulating the PE router transmits MPLS traffic at the specified rate. If frame loss occurs, the transmission rate is alternately reduced/increased using a binary search algorithm to determine the maximum rate at which the DUT can forward traffic without loss.

Results: Transmit rate, latency, data errors, sequence errors.


Figure 4. IxScriptMate L3 VPN RSVP-TE egress performance.

4. IxExplorer LDP Extended Martini Session Scalability Test

Objective: To determine the maximum number of LDP Extended Martini sessions the DUT can sustain at one time. Test setup: One Ixia port connected to the DUT (see Figure 5).

Methodology:

1. An OSPF session is established with the DUT.
2. An LDP basic session is established with the DUT, advertising the Ixia port loopback address as a FEC.
3. A number of LDP Extended Martini sessions (based on test expectations) are established with the DUT with one or more Layer 2 VC FECs advertised per session.
4. Status on the port is monitored to determine if all configured sessions are successfully established. Figure 8 shows the MPLS Martini labels learned by IxExplorer from the DUT one for each session.
5. If all sessions are successfully established, the number of extended sessions configured is increased and the test rerun. If not all are established, the number of sessions is lowered and the test repeated.

Results: LDP Extended Martini session capacity.


Figure 5. IxExplorer LDP Extended Martini session scalability test setup

Figure 6. Test results (learned labels)

5. IxScriptMate VPLS MAC Address Cache Capacity Test

Objective: To determine the maximum capacity of a router's Layer 2 MAC address cache.

Test setup: A minimum of three Ixia test ports connected to the DUT a Test port, a Monitoring port, and a Learning port.

Parameters: Number of PEs per port, number of VPLS instances per CE, number of hosts per VLAN, DUT MAC address table size, DUT MAC address aging time. (See Figure 7.)

Methodology:

1. OSPF and LDP are run based on user-configured parameters to set up the network topology and signal the VPLS instances.
2. Learning frames, equal in number to the specified DUT MAC address table size, are sent to the Learning port. These frames contain varying source MAC addresses and a fixed destination MAC address corresponding to the address connected to the Test port.
3. Validation frames are sent to the Test port back to the addresses learned on the Learning port.
4. The Monitoring port listens for flooded or mis-forwarded frames from the DUT.
5. A binary search algorithm is used to determine the maximum number of addresses that can be learned without flooding or dropping frames.

Results: Maximum MAC address capacity total and per VPLS instance. (See Figure 8.)


Figure 7. IxScriptMate run setup.


Figure 8. IxScriptMate test results.

6. IxExplorer RSVP-TE Fast Reroute Test

Objective: To measure the LSP switchover time of a router running RSVP-TE Fast Reroute between a primary LSP and a detour LSP upon link failure.

Test setup: Three Ixia test ports connected to the DUT one acting as the ingress and two as egress (primary and detour). See Figure 9.

Methodology:

1. Two bidirectional LSPs are signaled using RSVP-TE through the DUT from the Ixia ingress and egress ports.
2. RSVP-TE Fast Reroute objects are signaled to establish one LSP as a primary and the second LSP as a detour.
3. Unicast traffic is sent from the ingress port over the primary LSP.
4. A switchover to the detour LSP is initiated by bringing down the link to the DUT on the primary LSP.
5. Traffic is captured on both the egress ports. The switchover time is calculated by comparing the timestamp of the last packet received on the primary port with the timestamp of the first packet received on the detour port.

Results: Fast Reroute switchover time of the DUT.


Figure 9. IxScriptMate run setup.

Acknowledgements

Authors: Bruce Miller, Elliott Stewart

[ back | top of page | back to test plans ]