May 2017 | Reading time: ~9 minutes
In today’s enterprise environments, scaling voice infrastructure is no longer just about adding more phone lines. With Cisco Unified Communications Manager (CUCM), organizations are integrating voice, video, messaging, and mobility into a single, unified platform. This post examines how to effectively scale CUCM in complex environments, including clustering, media resource management, survivability, and licensing considerations.
Understanding CUCM Clustering
CUCM supports clustering of multiple servers to provide scalability and redundancy. A typical cluster includes a publisher and multiple subscriber nodes. To ensure high availability, servers should be geographically distributed and interconnected via high-speed WAN links with low latency.
Best practices include maintaining fewer than 80% CPU utilization under normal conditions and ensuring all nodes can reach the publisher for database replication. CUCM supports up to 20 nodes per cluster, with 8 call processing nodes.
Media Resources and MTPs
Media resources like Media Termination Points (MTP), Conference Bridges, and Transcoders are critical in VoIP environments. These resources can be hardware or software-based and need to be properly distributed across the network. Cisco IOS routers often serve as hardware media resource providers using the DSP resources.
Ensure that MRGLs (Media Resource Group Lists) are properly configured so endpoints and gateways can access required media resources. Centralizing these can lead to bottlenecks; distribute them across remote sites when possible.
Call Admission Control and Location Bandwidth Management
Scaling VoIP also requires careful planning around bandwidth. CUCM provides Location and Region settings that manage codec selection and bandwidth limits. Call Admission Control (CAC) helps to prevent oversubscription by denying calls when bandwidth thresholds are exceeded.
Use the RSVP Agent or Enhanced Location CAC (ELCAC) for more dynamic bandwidth controls, especially in WAN environments where video traffic coexists with voice.
Device Pools and Regions
Device Pools are used to group IP phones, gateways, and other devices with similar configurations. Assigning correct Regions ensures optimal codec selection between sites. Codec choices impact call quality and bandwidth usage—G.729 uses less bandwidth but lower quality than G.711.
Trunk and Gateway Considerations
As your enterprise grows, so does the complexity of call routing. Deploying multiple SIP trunks and H.323 gateways requires careful dial plan design and redundancy planning. Use Route Groups and Route Lists to prioritize outbound call paths, and configure fallback mechanisms for high availability.
Ensure that digit manipulation is handled consistently using translation patterns, calling search spaces, and route patterns.
Survivable Remote Site Telephony (SRST)
Remote sites depend on WAN connectivity to CUCM. When the WAN fails, SRST provides limited call processing locally using the site's Cisco IOS gateway. This ensures critical communications remain available even during outages. Configure SRST fallback and re-registration timers appropriately.
Licensing and Cisco Smart Licensing
CUCM licensing transitioned to Smart Licensing, requiring careful tracking of endpoints and features in use. Prioritize the use of Enterprise Agreement (EA) or Flex licensing for organizations scaling across multiple sites. Monitor licensing compliance through Cisco’s Smart Software Manager (SSM).
Monitoring and Optimization
CUCM provides RTMT (Real-Time Monitoring Tool) and CDR Analysis for troubleshooting and analytics. Scaling efforts should include proactive monitoring of call volumes, latency, jitter, and packet loss.
Network readiness assessments and periodic validation of QoS configurations are essential as call volumes increase and new services (like video) are added.
Conclusion
Scaling Cisco CUCM in an enterprise environment goes beyond adding users—it demands a structured approach to infrastructure, redundancy, resource management, and licensing. With careful planning and ongoing monitoring, organizations can deliver high-quality, reliable voice services to users across the globe.
No comments:
Post a Comment