How does multi-site clustering work?
NEC Clustering
Understanding and Selling the ClusterConcept
Sales Guide v1.1
Clustering Diagram
NOTE: To view, you will need Microsoft Visio 2003 installed on your computer. If you do not have this, download the Visio Viewer at the bottom of this page.
Objective
The purpose of this document is to explain in detail, the technical and sales implications of NEC clusters in a multi site environment. This document is targeted at a sales and pre-sales audience.
Introduction
When tackling a multi-site environment the traditional way of deploying voice, was to use a platform of the appropriate size at each location. These platforms would then all communicate with each other via a legacy protocol, such as DPNSS or Q sig. Using these protocols, or often a bespoke protocol designed by a specific manufacturer, the systems were able to communicate with a limited range of features.
Today with the advent of IP and the proliferation of reliable IP WAN services, the architecture for multi site scenarios, is changing both in the voice and data worlds. Where sites can be reliably and cost effectively linked, the concept of centralisation is becoming key, to any modern communications network. Where hardware can be centralised costs are dramatically reduced, administration becomes simple and future deployment becomes faster.
This type of architecture, of centralised hardware, has been a facet of the data world for many years. Now with the proliferation of VOIP the logical conclusion is that voice will follow the same architecture. This architecture has been termed by NEC and other manufacturers as “Clustering.”
Other Adopters of the Technology
Like NEC, other large communications companies such as Cisco have followed the same strategy. When Cisco launched its own voice product nearly five years ago, it naturally adopted the centralised architecture.
AVAYA, took a dramatic u-turn in 2001, to release its new range of Multivantage products. This product range was to adopt the centralised architecture and move the company from years of developing networked PBX technology.
The Key to Clustering
Within a multi-site environment, the key to clustering is to centralise hardware and thus reduce cost. However with centralised hardware, the risk of failure at local site becomes a real threat. This is overcome with the use of local redundant processing.
In simple terms this means all the users within the cluster use the processing power of the centralised hardware. Then if the connection to this hardware is lost, a local processing is provided by a local-redundant processor, keeping the local site alive.
The key benefit to this architecture, is that each site can be evaluated in terms of function and risk. Then local redundancy can be provided only at sites, where it is deemed essential the site survive in the event of a WAN failure. Sites not deemed to be essential, will be lost in the event of a failed connection to the central hardware.
Key Benefits to Clustering
- Allows expensive hardware only to be deployed to the sites where it is required.
- With all the clients using the centralised hardware, the platform acts as one single system across multiple sites
- Local redundant processing to provide survivability in the event of WAN failure or hardware failure at the central site
- Redundant centralised processing in the event of centralised hardware failure
- Administration is centralised with automatic updates to the outside nodes
- Growth of the platform is centralised allowing the roll out to integrate into the data network growth
- Deployment of application is centralised, negating the need for hardware at each site
- Overall system growth becomes more flexible as multiple clusters are linked
NEC USP
NEC’s number one USP is the migration story, which in a clustered environment is essential in making the architecture cost effective. At each site in the cluster, the NEC solution offer the customer total flexibility between the type of endpoint deployed, IP or traditional. Further to this the NEC offers full migration from a traditional digital endpoint to a full IP endpoint, with full retention of investment. All of this is achieved at a market leading price.
NEC Hardware
The NEC hardware is simple to design, configure and deploy. All cards are common at each site as are the handsets and licenses. The only item that changes is the type of CPU, an IPS always forms the central node of the cluster, with DMR’s where redundancy is required at the outside nodes.
Cluster Capacity
CPU Type Traditional Users IP Users
IPS Central CPU
500 Users at the central site* 500 Users at the central site*
DMR Local Redundant CPU 60 Traditional users at local site** 120 IP Users at local site**
* Numbers of users will vary depending on trunk configuration and number of IP/traditional users
** Number of users will vary depending on trunk configuration. The number of traditional users is available in conjunction with the 120 IP users.
Selling Clusters over Traditional Networking
With clustering it important to understand, which areas of the concept are going to appeal to the various parties within a perspective customer.
IT Manager/Director
- Ease of administration
- Ease of roll-out
- Control of the voice network in a format they can understand
- In the case of IP, the possibility of getting their network upgraded
Finance Director
- Reduced cost of ownership, in terms of admin and growth
- Reduced implementation cost through totally flexible migration (Unique to NEC)
Company Director
- Cost if of course the major concern
- Comfort factor buying the latest technology as support by other major vendor adoption
Cost of Ownership
Clustering often comes about through the need or interest for IP telephony, which is growing through most sectors of the market. Through this, comes the immediate need to upgrade data infrastructure to handle voice and data. This need to upgrade the data network, plays on the key NEC USP, flexibility and migration.
Flexibility
Customer sold on IP, centralisation and clustering can still achieve this architecture but with a mix of traditional and IP endpoints. Therefore, allowing the data infrastructure to be upgraded in line with its existing life cycle costing. By eliminating the need to remove and replace massive data infrastructures overnight, the customer is given the solution they require tomorrow at a cost they can afford today.
Migration
Flexibility is just one half of cost story, without migration the flexibility from day one is lost as the system grows. Through the retention of the handsets and their ability to become IP handsets, a large percentage of the initial investment is retained.
Summary
The NEC cluster offers the customer the latest architecture, whilst still growing with their data network, at the most competitive price in its market sector.
|