Management BYOD Infrastructure IoT Storage Security Privacy

Current Filter: >>>>>>

PREVIOUS

   Current Article ID:4921

NEXT



The elements of effective network monitoring

Editorial Type: Feature     Date: 11-2014    Views: 2231   









Networks are getting faster and more complex - but have network monitoring and diagnostics kept up? Rony Kay, CEO of cPacket Networks thinks not...

High-speed networks are at the heart of most businesses activities, and the ability to proactively monitor and mitigate imminent issues before they become costly disruptions is critical. As legacy network monitoring solutions weren't created to handle the complexity and scale of today's high speed data centre environments and cloud infrastructure, they are incapable of keeping up with the volume and speed of network traffic or distributed application environments, and not able to successfully deliver real-time operational visibility.

Modern computing infrastructure for distributed application depends on network connectivity between the applications and data centres. This emerging application delivery paradigm drives faster networks and a higher volume of traffic, while the user expectation for reliable and consistent quality of experience increases. Therefore the operational challenge of network monitoring and diagnostics becomes more difficult.

Unfortunately, legacy-monitoring approaches are not that proactive and they do not scale effectively to the growing challenges. For example, aggregating raw traffic from multiple links across the network and capturing vast amounts of packets on disk for post-analysis is a bottleneck by design. Analysing captured packets after-the-fact is not proactive and prevents the operational teams' ability to identify imminent issues as early as possible and take adequate corrective action ahead of costly disruption. The lack of real-time situational awareness handcuffs the operations teams and prevents them from proactively preventing negative impact on business activities.

Effective network monitoring and diagnostics needs to provide real-time granular performance indicators for specific application traffic across the entire network. It should rely on complete packet inspection at wire-speed and enable triggered interception of specific events based on pattern matching or abnormal traffic behaviours. Key performance indicators (KPIS), proactive alerts about anomalies, and interactive drill-down should be available from a unified dashboard for quick analysis, mitigation of issues and avoidance of substantial business losses.

Performing on-the-fly, real-time traffic inspection across the entire network environment by using distributed smart ports, will equip network operators with accurate information about bottlenecks, intermittent performance issues, and abnormal applications behaviours. It allows operators to firmly grip the network in real-time with pattern matching searches and set automatic alerts for undesirable error conditions. This enables operators to identify and correct issues before they negatively affect the end-user experience.

If operators have the tools to understand detailed network traffic behaviour, they can optimise performance in modern application delivery environments and data centres. For example, spikes and bursts are often the root cause of intermittent disruptions to various applications like media, interactive gaming, and automatic stock trading. Identifying these intermittent behaviours is of the utmost importance, but legacy monitoring solutions handcuff network operators by providing only coarse average link utilisation (over seconds or minutes) that obscures intermittent microbursts of application traffic. Not having this critical information is detrimental to capacity planning, traffic engineering and troubleshooting.

Operators with an integrated, unified dashboard of granular information from distributed smart ports across the environment, gain real-time visibility into high-resolution counters for any specific traffic profiles at the desired granularity. Having a broad situational awareness of KPIs across multiple hops in the path of the traffic enables the operations team to improve capacity planning, optimise traffic engineering, and shorten troubleshooting. This results in consistent application delivery with fewer disruptions and a better user experience and it avoids negative impact on business activities. These new solutions will radically simplify network monitoring and performance optimisation and provide the ability to troubleshoot in complex, high speed distributed network and data centre environments. Greater adoption of this new technology, with distributed smart ports, will be required with the ever more complex network infrastructure of the future. Legacy solutions are no longer fit for purpose.

Like this article? Click here to get the Newsletter and Magazine Free!

Email The Editor!         OR         Forward ArticleGo Top


PREVIOUS

                    


NEXT