Cross Talk: November 2009 Archives

Murdoch-Microsoft-The Dream Team, the "Killer App" is really Content or "It's the Content, Stupid." - Microsoft Delivers the News with News Corp

November 24, 2009 9:33 AM |

Continue reading Murdoch-Microsoft-The Dream Team, the "Killer App" is really Content or "It's the Content, Stupid." - Microsoft Delivers the News with News Corp .

"LinkedIn" Social-Business Network - Development Platform

November 23, 2009 11:42 AM |

0 Comments

Continue reading "LinkedIn" Social-Business Network - Development Platform .

"Presence" Super-Sized by OCS-SIP - Improves Critical Business Activities

November 22, 2009 10:58 AM |

0 Comments

Continue reading "Presence" Super-Sized by OCS-SIP - Improves Critical Business Activities.

OCS Forum Newsletter Today "Early Media & Late Media" Call Media Failures Bringing Chaos to SIP Networks

November 20, 2009 10:11 AM |

0 Comments

Click here for animated tutorial and explanation
Part 2 of 2 - Parallel & Sequential Forking with 1st & 3rd-Party Call Control with RCC

First Party Call Control - traditional telephony POTS, SIP and OCS are designed to provide for first party or first person call control.
Third Party Call Control - or third person call control is where another element, endpoint, server, telephone or device is involved in the call. Third party call control may mean that the endpoints share call control with another device such as a PBX, ACD-Automatic Call Distributor, CO-Central Office Switch or other device. The third party device such as a server may direct, redirect (fork) or disconnect the call.

NOTE: The animated tutorial explains the SIP signaling methods and potential "early media" clipping failures.
- Parallel Forking - the proxy forwards copies of the request to multiple destinations simultaneously.
- Sequential Forking - the proxy forwards copies of the request to one target at a time and
waits for a final response (or failure) before moving to the next address.

Key Point - Forking is critical to advanced SIP features such as "find-me follow-me." Critical to this process is ringing to let the caller alert the callee or (called party) of the call. In traditional PSTN-Public Switched Telephone Network communications "early media" refers to ringing and announcements to indicate the status of the call - ringing, busy, fast busy, call redirection, status "you have 12 callers ahead of you." In SIP, the forking process provides "early media" sending specialized ring-tones, audio announcements (e.g. call center status announcements), images or video before SIP session is accepted. However, there is NO common means of providing signaling to the receiver because of different types of hardware, softphones, UI-user interfaces, ringing devices and many other factors. In addition, early media may be omnidirectional or sequential forking, bidirectional or parallel-dual forking.

Early media failures can occur when the callee picks up and the UAS sends a 200 (OK) response with an answer, in parallel with the first media packets. If the first "early media" packets are received by the caller - UAC-User Agent Client, "media clipping (at the beginning of the media sequence)" or "media dipping" (at the end of the media sequence) can occur. This can occur is that the UAC cannot send media until the 200 (OK) response from the UAS arrives. Causes for clipping can be manyfold, however, UAC signaling, packet arrival delays), bandwidth, different SIP "methods" and commands between SIP systems and other factors. In addition, SIP signaling can typically take a different routing path than the media (communications) transmission which can be one of the factors causing media clipping.

"Late media" announcements (such as "will you take a survey" or (click for special offer") occur after the BYE and may have the same problems. Details can be found by reading RFC 3260.
In summary, incompatibility between SIP systems can result in communications chaos.

RCC-Remote Call Control also known as third-party call control is provided by CSTA-Computer Supported Telephony Applications. CSTA was developed by the European Computer Manufacturers Association (ECMA) and subsequently was formally standardized by the ITU-T, incorporating the Switch-to-Computer Applications Interface (SCAI).

CSTA is an OSI protocol stack that provides an open system interface to a PBX-Private Branch eXchange, ACD-Automatic Call Distributor or CO-Centrex central office switching. CSTA uses, among other technologies, SALT-Speech Application Language Tags specification and its SMEX-Simple Messaging Exchange element, telephony call control capabilities in MSS-Microsoft Speech Server to allow a developer to create sophisticated telephony-based speech applications that can exploit both basic call control services such as ANI-Automatic Number Identification (caller ID) and DNIS-Dialed Number Identification Service (800), using the included basic call controls, or extended call control services, to create custom call controls.

Updates Resource Center for Office Communications Server 2007 R2 and Clients

November 16, 2009 7:37 AM |

0 Comments

Continue reading Updates Resource Center for Office Communications Server 2007 R2 and Clients.

VPLS-Virtual Private LAN Service - Gloal Service for VPNS-Virtual Private Networks - Flash Animated Tutorial

November 15, 2009 1:12 PM |

0 Comments

VLPLS-Virtual Private LAN Service - Global Service for Virtual Private Networks

The animated presentation can be found at: http://www.techtionary.com/podcasts/vpls/

I originally wrote about VPLS sometime ago. I was prompted to update the information with an announcement from Verizon (see below). In addition, I wanted to remind network planners to have both redundant (multiple) and diverse (separate cable routes and trenches) both wireless and wireline services.

CIO Version - Here is the translation of those acronyms into "no-tech" English. Virtual Private LAN Service allows customers to create VLANs-Virtual Local Area Networks in a metro network or global network using IP-Internet Protocol MPLS-Multi-Protocol Label Switching or VPNS-Virtual Private Network Service. In CEO talk, this means you can connect all the branches in the same city without costing a lot and cities elsewhere "almost" as easily. In the animated presentation, you will see the differences between VPLS and VPNS. There are many others including training, availability, diversity, DR-Disaster recovery, application such as hub-spoke versus mesh networking and others. Check with your provider as YMMV-your mileage may vary.

CTO Version - LAN switching is a Layer 2 (Datalink) concept of switching rather than routing which takes place at Layer 3 (Network). In LAN switching all packets are "flooded" to all locations until they find their destination. Nice idea if you have only a few sites. Originally known as TLS-Transparent LAN-local Area Network Service, Metro-Ethernet, Ethernet Private Line and now often called VPLS-Virtual Private LAN Service, here are some reasons why you should consider this service.
First, it is easy-to-implement with plug-and-play installation.
Next, LAN Switches are cheap.
Third, VPLS provides very flexible BOD-bandwidth on demand options.
Fourth, it is a great solution for same-city customers with many locations.
Here are some reasons against: First, it doesn't grow or scale well to tens of sites and hundreds of users. Second, faster bandwidth doesn't provide QoS or solve flooding of packets due to large applications.

In Layer 3 routing, only the packets that are destined for the other location are sent. That is, packets are routed based on the destination IP address.
Here are some reasons for routing: First, it scales or grows essentially infinitely for customers with domestic and global locations.
Second, routing supports all kinds of IT systems new or legacy.
Third routing provides higher security.
Reasons against routing: First, there is a high cost with routing and routing will always be more expensive than switching.
Second, routing is complex to manage.
Third, routing requires very knowledgeable and technical staff.

I didn't forget VPLS, just needed a minute to get through the basics. VPLS uses MPLS to provide the "seamless" connections for VPLS. Shown in the presentation is the IP-Internet Protocol packet before and with the MPLS "label" attached or "tagged" on as it was originally called. MPLS consists of four elements, label bits, experimental bits, a stack bit and TTL-Time-To-Live bits which indicate the number of Label Switch Routers passed. Shown here is the "multi-protocol" part of MPLS and how it works with the other major networking protocols such as ATM, Frame Relay, Ethernet and others.

As shown in the presentation MPLS is used to connect locations and VLANS together.
In the next slide are the three key access topologies - ethernet-line for two sites, ethernet-LAN for three or more sites and ethernet-access for VPN/internet. This makes implementation and configuration of VLANS really simple.

WHY VPLS - Summary Here are a few business reasons: - Extends LAN infrastructure easily - Expands metro-net to global-net - Migrates Frame Relay to MPLS - Utilizes "pseudowire" approach - Needs "mesh" connections - Connect call centers via SIP - Provide DR-Disaster Recovery for data centers - Seeks greater customer (less carrier) control - Has multi- and broadcast traffic - Building a CDN-Content Delivery Net - Building an overlay multi-carrier network.

VPLS-Virtual Private LAN Service is going to be a great solution for any customer with more than location.

Here's Verizon's Information - Verizon Business Takes Virtual Private LAN Service Global
The following was provided by Verizon - "With this global expansion, the reach of Verizon VPLS will extend beyond the U.S. to customers in 31 international locations, including the United Kingdom, France, Germany, India, Australia, Canada, Hong Kong, Singapore and South Korea. Verizon Business leverages its extensive network of interconnection agreements with other in-region Ethernet providers to offer this expansive coverage.
Verizon VPLS, an advanced MPLS-based layer 2 virtual private network delivered via Verizon Business' innovative global-provisioning Ethernet platform, is ideal for organizations seeking to link multiple sites in a flexible, operationally efficient model. VPLS is protocol-independent, supporting both IP and non-IP applications, which allows customers to retain complete control of their own routing. In addition, VPLS allows customers to access Verizon's extensive portfolio of Ethernet services and other data networking services offered by the company.
In a challenging economy, Verizon Business' global expansion of VPLS service and growth of its Ethernet portfolio is a good-news story," said Stan Hubbard, senior analyst with Heavy Reading. The company's global provisioning platform for enabling a seamless experience, long history delivering Ethernet services, broad portfolio and stability will certainly all appeal to multinational customers.
VPLS is an optimal platform for converged networking as it allows for prioritization of applications via multiple classes of services (CoS). An organization can customize prioritization to ensure that business-critical applications receive the highest priority across the network. Verizon Real-time CoS, for example, makes VPLS ideal for carrying latency-sensitive traffic such as voice-over-IP (VoIP) and digital video. VPLS also enables customers to provision their network bandwidth from 1 megabit per second to gigabit per second and beyond, enabling efficient and cost-effective transmission of voice and data traffic. And VPLS is backed by robust service level agreements (SLAs), which customers expect from a carrier-class wide area network service." END - There was no direct link in the Verizon Press Release for more information.