Overview

One of the concepts that many junior voice engineers struggle with is the class of control features of Partitions and Calling Search Spaces. I remember feeling like I was banging my head against a wall repeatedly until I hit that Ah-hah moment and it all made sense.

Now there are many great explanations out there on the inter webs that are all technically correct and great. Some use the Lock + Key analogy like the CIPT1 Study guide and some use a White Pages/Directory Analogy. In my opinion, these can leave people without a whole lot of voice experience more confused than when they started.

What are Partitions and Calling Search Spaces used for anyway?

In my opinion, Partitions and Calling Search Spaces are primarily used as a way to find "things" in the phone system. They are used so endpoints (Phones, Gateways, Trunks) can dial resources (Directory Numbers, On-net Route Patterns, Off-net Route Patterns, Translation Patterns). From an end-user’s perspective these resources are generally represented by a number. Because you can use these Dial Plan constructs to limit what resources an endpoint can find, it allows you the ability to implement Class of Service.

Sure, you could just put everything in the none partition and be done with it all together but Tom Hollingsworth has already told us why this is a bad idea over at his blog networkingnerd.net

Yeah, ok. But you still haven't told me what Partitions and Calling Search Spaces do!

Hold your horses. This is the part you've all been waiting for....< Insert Drum Roll > Here it is:

•    Partitions contain "things"
•    Calling Search Spaces find "things"

That's it. I really believe this is the simplest way to think of how these two constructs work and behave.

Partitions

So we know that partitions can contain "things" but what exactly can they contain? A partition will only contain one or more of the following things:

•    Directory Numbers - These Directory Numbers represent other numbers in the Phone System that can be associated to Devices.
•    Route Patterns - Route Patterns represent off-net and on-net destinations and control how a call is routed to a destination. Route Patterns use Route Lists and Route Groups to send calls over a trunk or a gateway.
•    Translation Patterns - I consider Translations Patterns an intermediate step to either matching a Directory Number or a Route Pattern. They are used to Transform the Calling or Called Party Information before using a Calling Search Space to find a suitable match. It is also worth noting that a Translation Patten can also be set to Block instead of route which can be useful when using the Line/Device Approach for implementing Class of Service
•    Transformation Patterns - Transformation Patterns are similar to Translation Patterns except they are not part of the CUCM's routing construct. They are used for modifying Calling or Called Party Information for purposes of presentation but do not affect call routing. When you are using Transformation Patterns you should use dedicated Partitions and Calling Search Spaces for this function.

Calling Search Spaces

So if partitions contain "things" and Calling Search Spaces find "things", what do they look in too find them? Well they look in partitions, of course.

The following constructs use Calling Search Spaces to find things:

•    Devices - Phones use a CSS to find available patterns. This is generally referred to as the Device Calling Search Space
•    Directory Numbers - Directory Numbers also use a CSS to find available patterns. This is referred to as the Line Calling Search Space.
•    Trunks
•    Gateways
•    Translation Patterns

This list was by no means exhaustive and there is a vast number of features that require Calling Search Spaces to be defined to control access to DN's/Patterns for that feature. Things like Time of Day Routing, Presence, Call Forward Settings, etc

When a Calling Search Space is looking through a series of ordered partitions it doesn't work like an ACL in Cisco IOS that searches sequentially through the available partitions top-down until it finds a suitable match. No, It processes all listed partitions and matches on the best or longest match. The only time in which the order of the partitions matter is when you have equally specific patterns, in which the pattern in the partition closer to the top of the CSS list will win as a tie-breaker.

OK - So I think I get it, but I've noticed my Phone and my Directory Number both have a CSS. What's up with that?

Well CUCM is flexible in that you could choose to set a CSS on just one of these elements and leave the other blank.  When you have both a Device and a Line CSS the CUCM combines the two into a single CSS with Line CSS above the Device CSS.

There is a good reason to use both. In order to effectively use features like extension mobility but enforce local dialling habits at remote offices it is recommended to set the Device CSS as to allow full access to the Dial Plan and then use the Line CSS to restrict unauthorised patterns from being dialled. The Partitions in your Line CSS should contain Translation Patterns that are set to block the restricted patterns and also have the "Urgent Priority" flag checked (It is on by default).

If you would like more information about the Line/Device approach please check the SRND.

History

IPv4 is the addressing scheme that has been used on the public internet since 1981. IPv4 uses a 32-bit address represented as x.x.x.x and allows for 4,294,967,296 globally unique addresses.

In the early 1990’s it became apparent that the amount of addresses allowed for by IPv4 was not large enough to accommodate the continued growth of the internet. It looked at the time that IP addresses were going to be exhausted by approximately 1995/1996.

More intelligent use of addressing (CIDR) and Network Address Translation (NAT) eased the urgency of exhaustion greatly. NAT allowed individual networks on the internet to share a pool of global IP addresses, allowing the internet to continue growing without using anywhere near as many addresses. Even large business could have thousands of workstations sharing a single Global IP address to access the internet.

On the other hand, NAT is not the be all and end all as it breaks the end-to-end communication of the internet. The boxes that perform NAT can break many applications and cause issues for things like internet VoIP, peer-to-peer file transfers and disaster recovery. Significant development has gone into improving NAT technologies to the point they have become more robust and useable, however it still deployed as a duct tape style solution that goes against the original design features of the internet.

Nevertheless, we have finally reached a point where not even NAT can stop the addresses from eventually running out. The beginning of the end happened in February 2011 when the last 5 chunks of IPv4 address space (totalling over 81 million addresses) were allocated out to the five regional registries. APNIC - the IP address registry for the Asia Pacific region - was the first to run out of their IPv4 allocation in Mid-April 2011. This will give internet service providers around 12-24 months of reserved address space before we are completely exhausted.

What is IPv6

IPv6 is almost exactly the same as IPv4 but with a bigger address space. It uses 128-bit addressing and allows for 340 Undecillion globally unique addresses. To give you an idea this will allow each square millimetre on earth, including the oceans, to be assigned 67 million IP addresses each. 340 Undecillion addresses is enough for 1 IP address for every brain cell of every human who has ever existed on earth with addresses leftover for spare.

What can you do to prepare?

Unfortunately, the moment that all the IPv4 addresses are exhausted and you want to provide a new service out to the internet, you will have to use IPv6. If you don’t have IPv6 implemented across your networks, then you will not be able to access this service as it will only be available on IPv6.

Alternatively, if your organisation requires extra IP addresses for a new mail server or other external facing services, you will be in a spot of bother as IPv4 addresses won’t be available.

At the end of the day, everyone is going to have to make the switch sooner or later.

We strongly believe that it is much better to have a staged and well thought out IPv6 implementation than a rushed one when you realise you can’t access some website or service. Initial trial IPv6 implementations don’t have to be grotesquely expensive and will ensure any potential issues are identified and can be dealt with early – and not during a business-critical moment.

If you’d like to learn more about what your organisation needs to do to be IPv6-ready, email This e-mail address is being protected from spambots. You need JavaScript enabled to view it or call 6254 6600.

Modern building and campus developments in the Health Care industry benefit greatly from having an open, highly reliable and highly scalable communications infrastructure. When you view IP networks in the same light as other key utilities such as power, water and gas, you can leverage it as a key piece of infrastructure to provide end-to-end communications between telephony, server, building management, nurse call, HVAC, paging, MATV and CCTV systems.

These disparate systems – which in the past would need their own set of cable infrastructure, communications protocols and management stations – can now been seen as merely endpoints on a unified communications fabric. Instead of each of these systems being autonomous “islands” with no interconnectivity outside of its own closed loop, these systems can now be managed remotely and can be easily tied into other systems for improved interconnectivity, reliability and performance.

IP: An open and scalable platform

Utilising IP as your communications infrastructure allows you to benefit from a highly scalable system that can scale to millions of devices and spread across the globe while still providing a reliable level of service for all network endpoints and applications. As all IP communication is based on openly published and royalty-free standards, any system can be seamlessly integrated to utilise the IP infrastructure for end-to-end communication.

The IP network is completely agnostic to your choice of Building Management, HVAC or Nurse call system, yet is still able to provide transport for management, monitoring or interconnection into other systems, provided they can interface IP over any layer-2 medium such as Ethernet or 802.11 Wireless.

Convergence

The first major benefit realised from implementing a solid IP-based network infrastructure is the significant savings in cost from converging as many systems, messaging platforms and media types over a single communications platform. In this modern world, all buildings require high-performing and reliable network infrastructure to support data traffic, so by investing heavily in an enterprise grade solution, the network can be used to deploy a myriad of other applications and services over the same infrastructure.

Redundancy and reliablity

IP networks are highly fault-tolerant by design. By designing your network with redundant routers, switches and implementing multiple diverse traffic paths, the communications infrastructure being used by all applications and systems can guarantee constant and optimal service. Single points of failure can be completely eliminated from edge to edge of the IP network.

On an ASA you can test AAA Authentication from the command-line with the following syntax:

For example: