Architecture decisions for resiliency

The following are resiliency architecture decisions for the hybrid cloud network for classic infrastructure disaster recovery pattern.

Architecture decisions for Network Resiliency

Classic data center resiliency architecture decisions
Architecture decision	Requirement	Options	Decision	Rationale
High availability network path	Ensure availability of resources if outages occur. Support SLA targets for availability	Single IBM Cloud Direct Link Dedicated Single IBM Cloud Direct Link Connect Two IBM Cloud Direct Link Dedicated Two IBM Cloud Direct Link Connect	Two IBM Cloud Direct Link Connect	Two IBM Cloud Direct Link Connect provides a cost effective resilient solution with a short deployment interval and is flexible to meet both hybrid and multi-cloud strategies.
High availability gateway appliance	Ensure availability of infrastructure resources if outages occur. Support SLA targets for infrastructure availability	Deploy Gateway Appliance of choice in high availability pair	Deploy Gateway Appliance of choice in high availability pair	Ensures if one appliance is unavailable access is still available through remaining gateway appliance.
High availability path detection	Ensure the quickest traffic path recovery	Bidirectional Forwarding Detection (BFD)	Bidirectional Forwarding Detection (BFD)	Provides a much faster way of detecting link failures compared to the built-in mechanisms within routing protocols.

Classic data center disaster recovery architecture decisions
Architecture decision	Requirement	Options	Decision	Rationale
Disaster recovery network	Network disaster recovery capability in secondary region to meet recovery time objective (RTO) recovery point objective (RPO) requirements	Single IBM Cloud Direct Link Dedicated Single IBM Cloud Direct Link Connect Two IBM Cloud Direct Link Dedicated Two IBM Cloud Direct Link Connect	Single IBM Cloud Direct Link Connect	Provides a cost effective and flexible connection into a second region, with metered and unmetered billing options.
Global load balancing	Load balancing over the public network across two regions if there's an outage (DR) for failover to the other region.	Cloud Internet Services (CIS) Citrix Netscaler VPX DNS	Cloud Internet Services (CIS)	Provides a cost-effective solution for multi-region load balancing over public internet and offers extra security features