How to select a Network Management System

How to select a Network Management System

Introduction

With all the network monitoring tools available now, how do you find out which one’s the best for your network? Sure, cost should always be a key factor in the selection process, but you should always ask yourself, does this inexpensive NMS system covers your requirements?

No network is the same and of course different networks require different functions and features from a network management perspective. But in any case, there are some characteristics that are very important and can be applied to both Enterprise, Telecom and other networks management solutions. These can be considered as basic factors, besides the cost of the solution, that should be considered in the decision making process.

Which are they? Read below to understand!

The NMS requirements

Management of equipment and networks is required to be performed in a centralized manner.

In addition, local management, i.e. typically fulfilled by local craft terminals (LCT), must be using the same user interface and this should be web-based.

Management of multiple technologies from a single Network Management System is strongly required.

This will lead to the much needed elimination of vertical management silos.

NMS must be feature-rich, providing unified Fault & Performance Management and Service Provisioning for multi-technology networks.

Unified management is not only about unifying the user interface, as you will read below!

Aside from standard FCAPS functions, the NMS must also support advanced inventory, end-to-end service provisioning, auditing and historical performance monitoring.

This list can go on and on, but here the tip is the more functions supported the better your network monitoring becomes!

The NMS must support fine-grained roles & privileges for multiple users.

In case of dedicated NMS server, the Operating System must be hardened, to ensure compliance to strict NOC security guidelines.

NMS must provide a flexible and configurable user interface that uses tabs, context sensitive menus, dashboards and advanced drag n’ drop features.
NMS must support a multi-tier Client / Server architecture that fits small and large-scale networks.

Multiple processes must be able to run on a single Server, or be distributed to provide scalability and redundancy features.

NMS must support 24×7 operation without downtime during backups.

Failover and must be supported, at least with the capability of a redundant server with automatic data replication.

A big advantage of the NMS is the support of a High Availability (HA) Cluster setup, with hot standby, automatic switch over, and disaster recovery mechanisms.

New network elements must be able to be integrated in the NMS via software drivers or adapters.

New management features must be able to be added in the NMS with additional applications that share the common, fully integrated User Interface.

Support of open OSS / BSS integration protocols (XML / FTP / SNMP / SYSLOG / TACACS / LDAP / RADIUS) is a must.

XML or Web Services protocols is a big advantage.

NMS must be a full Java / J2EE implementation that allows using vendor-independent OS, Relational Database and hardware for lower CapEx & OpEx.

 

Why Unified Management?

The vertical management concept needs to evolve in order to overcome the numerous drawbacks identified, and this can only come from supporting unified management at the Element, Network and Service management layers.

When we are discussing unified management, these are the main concepts:

  •  Management of multiple technology domains from a single platform (wireless backhaul / transport, wireless broadband access, xDSL / VoIP access, etc.).
  • Unified user interface that is modern, simple and convenient to use.
  • Highly-effective service fulfillment & assurance to ensure improved user experience and reduced NOC costs.
  • Unified North Bound Interfaces (NBIs) thereby reducing the OSS integration workload.

 

Following the trend for network convergence, unified network management is required in various networking scenarios, thus simplifying network roll-out through centralized deployment and auto-discovery capabilities.

 

Spotting root causes of service unavailability

Service restoration time is critical for operator success.

Upon service unavailability e.g. at customer premises, alarms are typically issued by all the involved equipment. But, a service may involve multiple technologies, networks and devices, as is the case for home broadband. There, a service fulfillment process may involve routers, switches, DSLAMs, and customer premises modems.

Only a unified management system can quickly identify and rectify the root cause, which is responsible for services unavailability.

Alarm correlation in unified network management
2017-02-15T08:18:38+00:00 Jul 26th, 2013|Categories: NMS Essentials|Tags: |0 Comments

About the Author:

Christos Rizos is a Network, Systems and Applications Management expert with 20+ years experience in the technology domain, providing consulting services to Vendors and Operators around the world.

Leave A Comment