QuestionQ157

Network Design

A company runs critical VPC workloads that connect to an on-premises data center using two redundant, active-passive AWS Direct Connect connections. However, a recent outage of one Direct Connect connection showed that traffic takes more than a minute to fail over to the secondary Direct Connect connection. The company wants to decrease failover time from minutes to seconds.

Which solution provides the LARGEST reduction in BGP failover time?

  • A Reduce the BGP hold-down timer that is configured on the BGP sessions on the Direct Connect connection VIFs.
  • B Configure an Amazon CloudWatch alarm for the Direct Connect connection state to invoke an AWS Lambda function to fail over the traffic.
  • C Configure Bidirectional Forwarding Detection (BFD) on the Direct Connect connections on the AWS side.
  • D Configure Bidirectional Forwarding Detection (BFD) on the Direct Connect connections on the on-premises router.
Explanation

Bidirectional Forwarding Detection (BFD) detects forwarding-path failures much faster than BGP hold timers. For AWS Direct Connect virtual interfaces, asynchronous BFD is automatically enabled by AWS but takes effect only when it is configured on the customer router. AWS documents a 300-ms minimum BFD interval and a minimum multiplier of 3, enabling failure detection in seconds or less.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!