The flu seems to be making the rounds again. Typically, during flu season, the fix is just a quick shot in the arm and on our way. For those who opted out, it is now a game to avoid any downtime as we lead our non-stop, always on, hectic lives.
As consumers, we won’t settle for downtime either. The experience must be the best, all the time, every time. But, meeting these expectations is getting harder for network teams. Modern hosting and networking approaches, such as cloud and software-defined networking (SDN), are increasing complexity and decreasing visibility.
Network performance monitoringacross such dynamic and diverse infrastructures requires great teamwork. Just like the team at your doctor’s office, network operations need to be able to efficiently diagnosis and treat a problem in as few steps as possible. Shamus McGillicuddy form Enterprise Management Associates (EMA) agrees that:"When an IT organization implements SDN, network virtualization, and cloud networking technologies, it must carefully re-evaluate its existing network management tools. In doing this, many organizations will discover that their tools are not ready to support SDN-based network transformation."
Modern network performance monitoring should be designed to not only cope with complexity, but to control complexity. Not every outage gets the same fix. Instead of prescribing the same remedy every time, best-of-breed network monitoring should include pre-defined and guided workflows that can be adapted to reflect the make-up of both an organization’s network and IT teams.
"When an IT organization implements SDN, network virtualization, and cloud networking technologies, it must carefully re-evaluate its existing network management tools. In doing this, many organizations will discover that their tools are not ready to support SDN-based network transformation."
Thanks to these prescriptions for success, everyone in the performance value chain knows the remedy when encountering common scenarios, such as bandwidth utilization issues, devices not responding, or ‘slow’ services. This enables organizations to stay healthy when it comes to handling potential issues that could degrade performance.
Analytics also play a key role in increasing efficiency. By looking back at historical data, such as network flow, SNMP or SDx API data, recurring problems can be detected – and eliminated. Instead of wasting time with the same medicine that doesn't do its job, network operators and administrators can get to the root cause of problems faster. This not only safeguards the user experience, but maximizes networking resources.
Analytics, such as percentile reporting, baselining and projections, take network performance monitoring solutions from reactive to proactive. Dynamic thresholding and windowing help operators spot the difference between transient and real problems.
Analytics don’t just resolve current obstacles; they identify future opportunities. By embedding analytics into the service assurance lifecycle, organizations can monitor not only the performance of networking assets in real time, but also the quality of the end-user experience.
Jeremy Rossbach, Sr. Principal Product Marketing Manager at CA Technologies.
CA Technologies has launched a new network performance monitoring comic series featuring Kip and Gary to bring a little fun and laughter to the everyday challenges network operations teams face with traditional network performance monitoring along with implementing and managing modern network architectures like SDN and NFV. Check it out when you get a chance.