Disaster Recovery Planning for Business Continuity Session Name : Title
Introducing Jason Ouimette Product Manager, Noble Systems John Simpson CIO, Noble Systems Mike Mahfouz Director of Collection Operations, TeleCheck, TRS Recovery Services, Inc.
Glossary DR Disaster Recovery ROI Return on Investment RAS Report Archive Server VAS Voice Archive Server TTS Text to Speech R&D Research and Delivery BCP Business Continuity Plan
Disasters Happen! Wild Fires, Hurricanes, Earthquakes Health Epidemics Cut Framework at Local TelCo The Good News! We can provide the tools to be prepared
Understand Business Impact Estimate Financial Impact Loss of Revenue Loss of Customers Proportional to Severity of Disaster and Recovery Duration Demonstrate ROI Numerous DR Options DR Options can be similar to purchasing insurance. Imply costs
Standard Noble Redundant Items Hard Disks Raid 5 or Raid 10 Implementations Hot Swappable Disks Hot Spare Disk Only as good as last backup if multiple disks fail Power Supplies Redundant Power Supplies Plug into Multiple Power Sources Additional Items Redundant d NICs Spare Parts Kit available for all Noble hardware
Standard Topology
Noble Replication Server Licensed Noble Product Award Winning Product Large R&D effort to create Hardware and software Optional purchase item that can pay for itself in one event Replicates Primary Host Entirely Transaction Level Replication Replication processes are very efficient Failover process is complete within a few minutes Low Bandwidth usage
Noble Replication Server What is restored? System Configuration Data Agent Setup, Supervisor, Monitoring, etc. Campaign Data Calling Lists Composer Applications Custom Data Every critical part of the host system
Other Noble Systems Replication Offerings What other systems can be replicated? Telephony Servers/UTS Systems Replicates all IVR configuration Report Archival System (RAS) Voice and Video Archival System (VAS/VVAS) SIPhony Blade Systems
Noble Systems Replication Topology
High Availability Solutions
High Availability Solutions Composer Web Agent Can support the use of multiple web servers using clustering SIPhony Multiple SIPhony Blades Active Standby Systems
Composer Web Agent Use of Clustering Multiple Web Servers can be clustered together so if one web server fails, agents can continue to proceed. Failover to Active System If one host fails, Composer Web Agent can be configured to run off of an active host
SIPhony SIPhony Architecture SIPhony uses one main UTS (Unified Telephony Server) Many SIPhony blades are used to create what would once be one telephony box SIPhony Failure If one SIPhony blade goes down the system will continue to If one SIPhony blade goes down, the system will continue to function without downtime.
SIPhony Standard Topology
Active Standby System Architecture An active standby system can be created All Noble Systems would have an active standby server These servers could be synced at a specific time to try and keep some data integrity How does it help? If the primary active system goes down, Composer can switch to the other active system. This allows agents to log in without having a failover process. Inbound calls on SIPhony servers from failed active system can then be transferred to the online active system.
Testing the g DR Plan
Testing the DR Plan Document and publish the plan Distribute to Supervisors, IT, Facilities, and Agents. Prepare the run book Create a BCP Plan A business continuity plan suggests a more comprehensive approach to keeping productive. This would cover items such as: Staff Illness Departure of key employees Supply chain disruptions
Testing DR Plan (continued) Test the plan Run yearly test plan Simulate various scenarios Find weaknesses Update Periodically Business conditions change constantly Contact information of key personal may change Equipment location may change
Future Enhancements
Future Enhancements Database Clustering Database Load Balancing Creating a True Active/Active Solution
Q & A