Network management gets out of hand
Troubleshooting on the modern network has become a complex, time-consuming exercise. It's time to consolidate network monitoring.
As networks transform, network operations teams at major organisations around the world find themselves confronted with a complex, hybrid ecosystem comprising legacy and software-defined components, supported by siloed network monitoring and troubleshooting tools.
Today's monitoring strategies, consisting of multiple network monitoring tools, are simply not designed for scalable visibility into dynamic software-defined networks (SDNs), even though improved visibility and simplified operations are among the benefits network operations teams most hope to achieve from their network management technologies.
But because SDN creates agile networks that grow and shrink automatically in line with application usage, many major enterprises are deploying SD-WAN, or planning to do so soon.
However, the migration is a gradual one, and many enterprises will continue to run physical and virtual environments for some time to come. As organisations move to SD-WANs, many adopters are concerned about consolidation of their remote-site infrastructure, about automated traffic forwarding, and about the need for additional resources on site to manage new SD-WAN infrastructures.
In most cases, SD-WAN adopters must add yet another monitoring tool or outsource management of their networks in an effort to stay on top of this increasingly complex environment, with its webs of dependencies.
Network managers in half of all enterprises now have to grapple with 11 or more network tools simultaneously just to get a sense of what's happening.
Network managers in half of all enterprises now have to grapple with 11 or more network tools simultaneously just to get a sense of what's happening. They may even have to extract data from one system and manually enter it into another, to get an understanding of the environment.
Unfortunately, their best efforts seldom help them understand how conditions in one network element impact other elements within the environment.
At the same time, application performance has become crucial to business, and failure of a single application component can have costly ripple effects across the network.
Yet as enterprises embrace SaaS, PaaS and IaaS within their new hybrid environments, integration between traditional and cloud-based networks invariably leads to performance challenges. Simply put, enterprise applications are best suited for legacy LANs, and dependence on network reliability is directly proportional to the reliability of WAN delivery of enterprise-critical applications.
In a complex hybrid environment, most of the network manager's day is spent 'firefighting': reacting to issues that impact the performance of often crucial enterprise applications.
To get insight into whether there is room for improvement in their network management model, enterprises need to consider the following:
* Are we working within a fully integrated network infrastructure?
* What percentage of NOC staff time is spent responding to alarms?
* Do we have a clear view into all network activity: physical, virtual and logical?
* Are we able to clearly distinguish commonplace network events from those that require immediate action?
* Can we proactively identify all critical network events?
Most will find they are not getting optimal network management, and are wasting resources and risking application performance as a result.
The solution lies in intelligent consolidation and simplification, or 'NetOps to the Power of One'.
The next-generation network monitoring solution must address the complexities and inter-dependencies across hybrid environments, spanning protocols and vendor landscapes, and visualising the relationships among the network elements, with a single dashboard that supports traditional, software-defined and virtual networking in a single context.
It must give full-stack visibility and eliminate network blind spots introduced by cloud workload migrations, deliver open APIs to integrate with business intelligence dashboards and reporting. It must even go a step further: actually harnessing advanced analytics to constantly search for optimisation opportunities and proactively identify potential problems across traffic, flow, trends, performance, systems and apps.
By tying network monitoring back to an analytics engine, it becomes possible to gain operation intelligence that actually anticipates the possible impact on the user experience itself. And because applications are the heart of the modern enterprise performance, the modern network monitoring solution must take a highly application-centric approach.
With one converged and customisable view of network operations, network managers can perform comprehensive and scalable monitoring and analytics.
And with a single context for troubleshooting application experience issues related to network impact, network operations staff are able to focus on what's important by simplifying root cause analysis.
Enterprise Management Associates research found those who use one to three network monitoring and troubleshooting tools on average catch 73% of network problems before users detect and report them.
But organisations that use 11 or more network monitoring tools typically catch only 48% of network problems, and 37% of those deal with several outages a day.
By reducing the number of solutions at play, consolidating management under a single pane of glass, and deploying next-generation analytics to proactively optimise the network, the network operator can deliver dramatically improved network performance, significantly faster time to resolution and improved user experience. For the enterprise, this means better productivity and enhanced customer service.
Business unit lead, Agile Operations, CA Southern Africa.
Miles Bowker joined CA in November 2012 in the role of senior architect and progressed to the position of pre-sales manager and more recently business unit lead of Agile Operations.
Prior to this, Bowker studied computer engineering and then went abroad to gain business experience in Europe.
At CA Southern Africa, he is responsible for the building and delivery of Agile Operations solutions in line with sales targets and strategies. He coordinates activities across teams, both internally and customer-based, with the goal of ensuring the client's business objectives are achieved.
Miles Bowker joined CA in November 2012 in the role of senior architect and progressed to the position of pre-sales manager and more recently business unit lead of Agile Operations. Prior to this, Bowker studied computer engineering and then went abroad to gain business experience in Europe. At CA Southern Africa, he is responsible for the building and delivery of Agile Operations solutions in line with sales targets and strategies. He coordinates activities across teams, both internally and customer-based, with the goal of ensuring the client's business objectives are achieved.