Webex Calling Local Gateway Sizer

Three reasons to deploy a Local Gateway in Webex Calling: (1) You already own a SIP trunk from a carrier not on Cisco's CCP list. (2) You have regulatory requirement to keep PSTN breakout local (some EU jurisdictions). (3) You're migrating from on-prem CUCM incrementally and want a single trunk that serves both old and new platforms. If none of these apply, Cisco Calling Plans or CCP (Cloud Connected Provider) is simpler — no router to manage.

CUBE sizing is driven by concurrent calls (CCS), not users. A 300-user office at 6:1 ratio needs 50 CCS. A contact center of 100 agents at 3:1 needs 33 CCS but with much higher PPS (packet per second) from busy-hour traffic. The same router model can handle both, but you need to enable the right IOS-XE package (uck9, securityk9). Concurrent calls = 2 × user-side + 1 × carrier-side on CUBE, because it has two legs.

Transcoding is where DSP farms come in. If Webex is G.711 and your carrier is G.729, every call needs transcoding — and DSP resources are fixed per card. A PVDM4-64 handles 64 G.711↔G.729 calls. If you don't have transcoding, skip the DSP entirely; CUBE can pass-through codecs natively. Opus at the Webex side and G.711 at the carrier side does require transcoding.