Welcome to the Cisco Networking Professionals Ask the Expert conversation. This is an opportunity to get an update on how to design CVP based Cisco Contact Center enterprise solutions. Shahzad Ali is a technical marketing engineer with the Unified Communication Systems Architect team at Cisco, where he is responsible for system-level network designs and strategy for Cisco Unified Contact Center Enterprise and Unified Customer Voice Portal products. He is also involved in writing solution reference network designs and documenting best practices for solution designs. Shahzad holds a double CCIE certification (#10650) in Voice and Routing & Switching. He holds a bachelor's degree in computer systems from NED University of Engineering and Technology, Karachi, Pakistan, and a master's degree in electrical engineering from Wichita State University.
Remember to use the rating system to let Shahzad know if you have received an adequate response.
Shahzad might not be able to answer each question due to the volume expected during this event. Our moderators will post many of the unanswered questions in other discussion forums shortly after the event. This event lasts through September 5, 2008. Visit this forum often to view responses to your questions and the questions of other community members.
We want to terminate SIP trunk directly to the CVP Call Server without putting the Cisco Unified Border Element (CUBE) in the middle. What steps do we need to take to make it work?
This is really a good question and in order to successfully design a CVP based Contact Center it is important to know the details about it.
The Cisco Unified Border Element (CUBE) serves as a feature-rich demarcation point for connecting enterprises to service providers over IP voice trunks.
When connecting to a third-party SIP device, inluding a SIP PSTN service provider, if CUBE is not placed between CVP and third party SIP device, the customer or partner is responsible for doing the integration testing to ensure that both sides are compatible.
Also, you have to keep in mind that when connecting to a PSTN SIP trunking service without CUBE, how the connection will be secured. And how NAT and/or address hiding will be accomplished. Otherwise it is possible for the service provider network to have full access to the customer network.
CUBE addresses both of these concerns and it is the service-provider interconnect interface recommended by Cisco.
We own Nuance NVP System that is the backbone of the solution. Nuance NVP consist of 1-Nuance Telephony Server which provides SIP Services. 2- Nuance Conversation Server which provides ASR/TTS services 3-Nuance Voice Server which is basically a VXML server. We wanted to know if Nuance VXML server will be supported by the Cisco IOS voice browser and what could be other caveats we should be looking into?
Since VXML is the W3C's standard XML format and our IOS Voice Browser (VXML Gateway like 3800, 5400 etc.) supports it, this solution should work. For support issues, the TAC engineer will do due diligence to determine if Cisco components are working as per standard or not. For the ASR/TTS component, the Business Units within Cisco has done the testing with the Nuance. Please take a look at the following URL for the Cisco IOS Compatibility Matrix with various AST/TTS applications
But you should be aware that the testing was done with the CVP VXML Server and CVP CallStudio and this is what we approved. And all the numbers are being published based on it. In CVP SRND you will find a note saying...
"These performance numbers are accurate when used with either the Cisco Call Server or Cisco VoiceXML Server. Performance can, and often does, vary with different applications. Performance from external VoiceXML applications (such as Nuance OSDMs) might not be representative of the performance when interoperating with non-Cisco applications. You must ensure that the CPU usage is less than 70% on average and that adequate memory is available on Cisco gateways at full load when running external VoiceXML applications. Users should contact the application provider of the desired VoiceXML application for performance and availability information. Be aware that external VoiceXML applications are not provided by Cisco, and Cisco makes no claims or warranties regarding the performance, stability, or feature capabilities of the application when interoperating in a Cisco environment."
In your case, you have to do the testing yourself with the Nuance VXML Server (Or Nuance Voice Server) and figure out the performance number like number of VXML sessions etc.
Another area to be aware of is the Call Survivability. If you don't use CVP in your solution then you can't fully use a great feature called Call Survivability. For Unified CVP calls, survivability is handled by a combination of services from a TCL script (survivability.tcl) and SRST functions. The survivability TCL script is used to monitor the H.225 or SIP connection for all calls that ingress through the gateway. If a signaling failure occurs, the TCL script takes control of the call and redirects it to a configurable destination or plays generic prompt or busy tone. The destination choices for the TCL script are configured as parameters in the Cisco IOS Gateway configuration.
Is the Java Source code for the Web Services Integration element available? We are having an issue with how the element is handling empty strings. We think we can fix it if we had the source code for the web services element and we could build a mechanism for handling empty strings. The element is throwing a null exception error every time we get an empty string back from our back end system?
This question doesn't relate to the CVP design dicussion that we are having here it is more like a CVP Development related question to me. But I will try to answer this for you.
The quick answer is that Cisco does not release source code fo this element.
But having said that, if you think it is a defect, you can open a TAC ticket. And this way engineering can fix the issue depending on the severity of the defect.
Also, I would highly recommend you to consider purchasing a Developer Services Contract. Developer Services for CVP allows you to open a TAC service request and get more advanced answers and help beyond what TAC can covers.
One more option, and this is reall interesting and useful. Cisco recently launched a community based developer forum @ http://developer.cisco.com. It is the forum for all the developers (not just cisco) to come and join and post development related questions like you have.
There are developer fourums like
* Customer Voice Portal (CVP)
* Enterprise Contact Center Scripting (ECCS)
* Contact Center Enterprise/Hosted/ICM IVR PG (GED-125)
* And many more
I hope this would help.
The Ops Console is a convenience - it was not there in CVP 3.x. It was added to the CVP 4.0 line to make things easy to manage in a complex, distributed environment containing a number of moving parts.
It's a nice tool, and I would not want to be without it. For example, the CVP 3.x way of pushing a big Audium application from the Studio box to the VXML server over FTP was extremely slow; now saving it to the Studio as a ZIP archive and using OAMP to push it to the VXML server(s) is way faster.
But there is nothing it can do that you cannot do another way.
No. Since OAMP Server component does not involve in any call processing so even if goes down it should not affect call processing.
The Operations Console Server provides an Operations Console for the browser-based administration and configuration for all Unified CVP product components.
Just for reference following URL would be handy
I have a very simple straight forward question.
When should we propose CVP / IPIVR (CRS)?
I will give you a brief comparision between CVP and IP-IVR that should help you propose a solution
The biggest advantage of having CVP vs IP-IVR is that with the CVP you can queue calls at the edge gateway while the caller is waiting for an agent to become available where as with the IP-IVR (CRS) you cannot achieve that.
CVP uses VXML and its queuing at the edge gives you significant advantage for WAN bandwidth saving. The prompts are stored in the gateway flash and played to the PSTN caller. If prompts need to be fetched from the medial server, they will be fetched and then cached in the GW for subsiquent use
There are others factors that you can consider. For example
CVP 7.0 supports 500 H.323 OR 850 SIP ports per server vs. IP-IVR support 300 ports
The Scripting Tool for CVP is CVP Studio (Eclipse-Based) that can be used to write complex and sophisticated VXML applications where as IP-IVR has its own Editor to write scripts.
CVP supports both centralized and distributed deployment models vs. IP-IVR supports centralized deployment models only
With CVP (as mentioned before) the medial is terminated on the Voice Gateway vs. in IP-IVR media is terminated on the IP-IVR server.
CVP supports any mix of codec on the gateway vs. IP-IVR supports either G.711 or G.729 (that is set at the time of installation)
If you want to have ASR/TTS support then both will support it but with CVP MRCP interface at the voice gateway where as with IP-IVR MRCP interface on the IP-IVR server.
These are the few that I mentioned here. Please let me know if you have any other question or wanted to know any thing more in detail and I will be happy to assist.
I am relatively new to this platform, I hope this is the right way to ask the expert.
My question, in short, is, how do i make outgoing calls from IPCC? What i want is a subscriber to call into my contact center and the script automatically calls another system and connects him. The detailed question is on here but i have not received a response yet.
Thanks very much!
I can only give you a high level overview here. But basically the logic is that you can use transfer option in the script to transfer incoming call to any other location which could be a ACD or another contact center.
Following link could give you some clue as well. There are different ways of achieving the same.