With all of the choices and complexities out there, how can organizations create the best business continuity strategy?

 

In Andrew Prokop's recent No Jitter article, "Peeling Back the SIP resiliency layers," he addressed some of the methods available to increase redundancy when implementing SIP solutions. I'd like to take this topic a step further and dive into the intricacies of business continuity for unified communications.

 

When implementing a unified communications platform in a multisite environment, designing for business continuity is an imperative for most organizations. Balancing the cost of a highly available solution against business requirements while understanding where the various redundancy options fit, is a challenge that plagues many ICT departments.

 

Carriers, UC components, data network components and virtual server hardware all offer redundancy alternatives. With all of the choices and complexities, how can organizations create the best business continuity strategy? Should they focus on one component or include components from all disciplines? Is there one best solution for all organizations? Or even one solution for all divisions and locations of the same organization?

 

Cost is often a primary factor in considering redundancy alternatives, and cost must be weighed against the business impact of a disruption.

 

The first step in designing a business continuity strategy is to determine which elements, locations, divisions or services are most critical to the organization.

 

For example, is there a critical contact center that generates a large portion of the company's revenue? If so, are all of the agents at one location, distributed across multiple locations, or remote workers? In a configuration where contact center agents are highly critical to the company's bottom line, the following strategies might make sense:

• Hardening the contact center application

• Sourcing from more than one server and more than one data center

• Ensuring incoming calls can traverse trunks from different carriers, through redundant SBCs and different gateways, and via different physical entry points

• Including the option of failover to VPN connectivity, while utilizing best practices for ensuring LAN resiliency

 

In a case where you have a remote worker in a nonessential warehouse, for example, spending a large amount of money to provide a fully redundant, fail-proof design might not represent the best use of corporate funds. The correct business continuity strategy for such a worker might be only to provide a 15 minute UPS for the switch, or no wired redundancy at all. If the MPLS fails, it might make the most sense for that worker to simply use a cell phone.

 

Designing a best-of-breed business continuity solution, then, requires not only knowledge of the design alternatives available, but a firm grasp of the true business needs of each location being included in the design. Ideally, this information should already be available, since understanding business requirements and use cases for different groups in the organization is a necessary precursor to a well adopted unified communications solution.

 

Design Strategy

When thinking about business continuity requirements, some considerations are:

• How critical is the location or group to core business needs?

• How much revenue will be lost if this group is unable to communicate?

• Would there be a life safety issue if the group is unable to communicate?

• Would a city, state, federal regulation be violated if this group is unable to communicate?

• Is voice communication critical to this group? Would email or chat be an acceptable alternative if voice communications are not available?

• Will this group continue to function if communications still work but there is no power in the building? No air conditioning? No heat? No lights?

  • For 5 minutes?

  • For 30 minutes?

  • For an hour?

  • For a day?

  • For multiple days?

• If there is a fire or natural disaster that prevents this group from accessing their office, do they still need to communicate?

• If the group is able to access the building and there is power in the building, what happens if the network goes down? Or the SIP trunks? Or the core communications server? Or the DHCP server? Or the LAN? 

 

Once the business needs have been identified, determine which design alternatives best meet specific business continuity requirements.

 

Carrier Services Business Continuity Strategies

• Use different carriers (inbound traffic and toll free numbers will inherently be with one carrier)

  • Use different carriers for MPLS and SIP 

  • Use different last mile providers for local loops

• Use carrier offered failover -- some, but not all, carriers allow failover from SIP to PRI, POTs or wireless device, for example

• Ensure MPLS, SIP and Internet circuits do not share the same Ethernet circuit

• Ensure different physical routes and conduits for outside cable plant into buildings

• Utilize carrier failover options such as reroute in the event of a SIP or MPLS failure

• Implement redundant SBCs for SIP connections

• Implement redundant gateways/routers for MPLS, Ethernet, and PRI 

• Utilize failover to VPN over Internet as an alternative route for remote locations

• Utilize LTE backup at both head end and remote site for alternative routes

• Distribute trunks to multiple locations, rather than centralizing trunks

 

Equipment Manufacturer Business Continuity Strategies

• Deploy UPS or battery backup for critical common servers and critical distributed servers

• Use backup generators for long-term back up of critical components

• Utilize a minimum of N+1 for critical core components 

• Use N+1 or greater for high quantity components

• Distribute resources such as voice mail, contact center, IVR, call recording, etc. to one or more remote locations

• Distribute skilled staff so that critical operations are not dependent on the survival of one location

 

Network and Data Center Business Continuity Strategies

• Use multiple, hardened data centers

• Use geographically diverse core locations

• Leverage virtualization to provide alternative resources for core server functions

• Distribute network resources such as Active Directory and DCHP servers

• Employ other best practices for data network resiliency

 

Analyzing specific business continuity needs should be part of your overall needs assessment. Assessing needs by group or location and planning at a granular, rather than global level, will ensure survivability while helping to contain spend normally associated with high level availability. By taking a holistic approach to implementing disaster recovery, businesses can assure the solution implemented meets the needs unique to their organization. 

 

 

"SCTC Perspectives" is written by members of the Society of Communications Technology Consultants, an international organization of independent information and communication technology professionals serving clients in all business sectors and government worldwide.