Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Exhaustive Failure Simulation

Use Routing Director to simulate an exhaustive failure scenario in your network model. A report is generated after you run the simulation. Check the reports to understand the impact of various failures to the network.

The exhaustive failure simulation automatically fails each and every link, node or SRLGs within the network one at a time, without the control over the selection of elements.

To create an exhaustive failure scenario:

  1. Navigate to Planning > Networks > Offline Models.
    The Working Models page is displayed.
  2. From the list of network models (offline version of a live network), select a network model on which you want to simulate a failure scenario.
  3. Click the Simulation button to define a failure scenario that you want to apply to your network model. The Exhaustive Failure Simulation page is displayed.
    Note:

    Before you run a simulation, ensure that you save the changes made to the network model by clicking More > Save Model.
  4. Complete the configuration according to the guidelines in Table 1.
    Note: Fields marked with an asterisk (*) are mandatory.
  5. Click Simulate to create a failure scenario.

    A message indicating that the simulation is complete is displayed. Depending on the size of the network the simulation task might take longer to complete, despite the message indicating completion of the simulation. Reports are generated only after the simulation is complete.

  6. On the Reports page (Working Models > select the model > click Reports), check the reports to understand how your network reacted to the failure scenario. For more information, see View Reports.
Table 1: Fields on the Exhaustive Failure Simulation Page
Field Description
Name

Displays the name of the selected working model.

Note: You cannot edit this field.
Layer*

Select the layer for which you want to create the failure scenario. You can choose one of the following:

  • Demand (Demand bandwidth)

  • Tunnel (Tunnel bandwidth)
Description Describe the failure scenario that you are creating.
Exhaustive Failure*

Select the exhaustive failure mechanism you want to simulate.

  • Single—Checks the network impact when one of the links, nodes, or SRLGs in the selected failure type fails at a time.

    For example, in a network with 10 links, Routing Director fails one link at a time and simulates how the network responds. The same applies to node or SRLG failures.

  • Double—Checks the network impact when any two or a combination of selected links, nodes, or SRLGs fail.

    If you select Link and Node as the failure types, then Routing Director simulates failure of below combination of nodes and links and its effect on the network:

    • Two nodes

    • Two links

    • One node and one link

    For example, if a network has 3 nodes and 3 links, the double failure simulation will include:

    • 3 scenarios of 2 nodes failures

    • 3 scenarios of 2 links failures

    • 9 scenarios of 1 node and 1 link failure

    In total, you obtain number of failure scenarios as 15.
Failure Type*

Select one or more failure types.

You can simulate failure scenarios for links, nodes, and Shared Risk Link Groups (SRLGs).

The simulation is performed to assess how the network is affected when one of the links, nodes, or SRLGs in the network fails at a time.

For example, if a network has 10 links, at any point in time, Routing Director fails only one link at a time, exhaustively and iterating through all 10 links, and simulates how the network would react to each failure scenario. The same is applicable for the node or SRLG failure.
Number of Instances

Select the number of instances for elastic simulation optimization

  • 1—Routing Director runs simulation using a single instance.

  • 2 through 5—Routing Director runs simulation using multiple instances in parallel. That is, elastic simulation optimization is enabled.

Elastic simulation optimization distributes jobs across multiple instances and executes these instances in parallel. An instance is a processing unit allocated by Routing Director to run simulations independently. You can increase the number of instances to speed up failure scenario execution, therefore reducing the time needed to execute failure scenarios.

By default, Routing Director runs simulations using a single instance. When you select the number of instances from 2 through 5, Routing Director distributes the total simulation jobs across the selected instances.

For example, in a network with 10 links and 2 instances, Routing Director divides the 10 links between the two instances. Each instance then concurrently fails one link at a time and simulates how the network responds. The same is applicable for the node or SRLG failure. This parallel processing reduces overall execution time.

Note: A maximum of 5 instances (slots) are available to run exhaustive failure simulations in parallel. If all instances are in use and if you start a new simulation, Routing Director places the new simulation in a queue and starts it only after instances become available.

For example, if one failure scenario uses 3 instances and is still running, and you start another scenario that requires 5 instances, Routing Director does not start the second simulation immediately. Instead, it adds it to the queue and starts it only when enough instances become available.

Related Documentation