by Stuart Hardy, Business Unit Manager of EOH’s Carrier and Network Solutions Division

Part 1

A network monitoring system has become a strategic property of cloud environments, and its fundamental property is resilience. The main reasons for network monitoring are to maintain the network’s health, ensure availability and improve performance.

More specifically, autonomic network monitoring – the ability for a network to self-manage its distributed resources by automatically reacting to unpredictable changes – is an essential requirement for cloud networks.

This is not a trivial undertaking. It requires the implementation of a control loop that receives inputs from a vast number of sensors (the monitoring data) and propagates control actions to a large number of distributed actuators.

The requirements of effective cloud network monitoring

Network monitoring is far more strategic than its name implies. It’s not only about constantly checking for problems; it’s also about optimising data flow and access in a complex and changing environment.

A cloud network is complex in structure. It consists of several layers to allow for functional separation, modularity and thus manageability. However, such strong layering poses several limits on the monitoring system, and the kinds of analysis and consequent actions that can be performed.

Consumers and providers therefore need to make decisions based on a limited horizon. Overcoming this limitation is very challenging in terms of technology, privacy and administration.

Effective monitoring techniques should be able to provide both very fine grained measures and a synthetic overview look at the cloud, involving all the variables affecting the QoS and other requirements.

They should be integrated with a control methodology that manages the performance of the enterprise system.

And, more importantly, the monitoring techniques should not add a performance burden to the system.

What to monitor

Deciding what to monitor on your network is as important as monitoring itself. To make the right decisions in this regard it is necessary to ensure that the network topology map is up to date. This map should accurately lay out the different types of networks to be monitored, which servers are running which applications on which operating systems, how many desktops need to be included and what kind of remote devices have access for each network.

Gaining clarity on these things at the outset makes it simpler to choose the right monitoring tools.

Other usual areas of monitoring include: bandwidth usage, traffic, application performance and server performance, as well as the performance and health of network routers and switches.

Long-term analysis, trends and peak loads can be obtained through extensive network monitoring evaluations, and resource availability can be planned according to demand. This is necessary to guarantee consistent IT performance across virtualised systems.

A private cloud will only function smoothly if a fast, reliable network connects the physical servers. For this reason, the entire network infrastructure must be analysed in detail before setting up a cloud environment. To ensure that users have constant access to remote business applications, the performance of the connection to the cloud must be monitored on every level and from every perspective.

You will also want to identify specific activities and performance metrics that produce results which enable you to address specific business needs, such as meeting compliance requirements, removing internal security threats and providing more operational visibility. And you can build a database of critical information that you can use to plan for future growth.

Network monitoring outcomes

A network monitoring system will help make sense of these complex environments, providing reports that managers can use to:

• Confirm regulatory and policy compliance
• Highlight potential cost savings
• Solve efficiency-sapping mysteries like dropped mail sessions
• Help determine employee productivity
• Spot overloaded equipment before it can bring down a network
• Identify weak WAN links and other bottlenecks
• Measure latency, or the delayed transfer of data
• Find anomalous internal traffic that might be indicate a security threat


Stuart Hardy has been in the ICT industry since 1997, has been in the Telecommunications industry since 1997, intimately involved in product development, operations and product marketing roles. He has held Executive level positions in some of the largest Operators in South Africa and has founded and driven two successful start-up companies in the Mobile data and Wireless networking spaces. Today, Stuart is a Divisional Director for EOH in their Telecommunications sector.


by Stuart Hardy, Business Unit Manager of EOH’s Carrier and Network Solutions Division