Hi,

I will give 2 examples: 1 for

Translation Pattern use, 1 for CTI Route Point and my recommendation at the end.

Example A) Translation Pattern:

Simple translation pattern: 1XXX

Adding a translation pattern will populate our Digit Analysis (DA) Engine within ccm process with the pattern 1XXX in the specified partition.

When Device A dials the number 1XXX and hits this pattern a high level overview of what exactly happens will be the following:

Find the best match in our Dialing Forest, match Translation pattern and route the call to the next destination in place (a Phone, a route pattern, another Translation pattern, etc) until

we find the final destination or we fail to reach it.

A low level overview will be:

1. Device Layer, Call control (cc) layer (IP Phone, Gateway, Trunk, etc) obtains digits and passed them to DA

2. Once in DA, DA will try to match the digit string. If digits string is not match DA will send a block response. cc child process (per this call) will send instructions to release the call, reorder Tone will be played.

3. If digit string matches a valid pattern, DA will send a request to DeviceManager asking about CEPN (callable endpoint name) This is a pkid in DB. The signal that DA send to DeviceManager is a DMPidREQ

4. If DeviceManager finds the CEPN, then it returns the SDL pid of the endpoint

process in the DMPidRes. If the CEPN is not found then DeviceManager returns a DMPidErr

5. Assuming the CEPN was found, the SDL pid will be sent to DA and then from DA to Child Call control in the DAres signal Cdcc will then send a CCSetupReq signal to the endpoint process.

6. Call is extended to desired destination (more stuff hapens here) but we ring a phone somewhere else

Example B) CTI Route Point,

CTI Route Point with Directory number: 1XXX

Adding a CTI Route Point will populate our Digit analysis Engine with the pattern 1XXX in the specified partition.

Will add another Device to the Device Manager process and will update our Forwarding Tables to the CFA, CFB or CFNA destination entered at DN level.

In case we have this CTI Route Point associated to a user monitoring this device via CTI (CTI controlled) CTI Manager service will take control despite the Device is Out of service.

When Device A dials the number 1XXX and hits this pattern the process will be the same as above for translation pattern except that we are going to introduce more delay because of an intermediate process: Forwarding Layer

This will happen at the end of 6 and will cause the call to go back to point 1.

We know that under normal conditions a CTI Route is a valid destination device, but in this case this Device is unregistered and Out of service (OOS).

This will cause to follow Call Forward rules and send the call to pattern specified in the Call forward settings, introducing an extra process being invoked and duplicating the route process

I would recommend Translation patterns over CTI Route points, because of the following:

- CTI Route Point original design is not as Transformation numbers, is for redirection to CTI controlled devices.

- Ease of management:: CTI Point introduces extra points of failure, (Forwarding Process, Directory number settings ,Call forward Settings, Device Settings)

- The use of CTI Route point and Call Forward configuration involves the use of CF Calling Search spaces which makes troubleshooting more difficult when multiple partitions and devices are involved in the call.

- The use of Translation patterns allow us the addition/deletion via AXL with less parameters than the ones for CTI Route Points.

- The use of CTI affects the calculation of BHCA for cluster which is using CTI controlled application.

http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/srnd/7x/callpros.html#wp1084384

- Extra CDR record information may be generated with the use of CTI route Points since we have a redirection in place.