By: Doug Roberts, VP & GM – Enterprise and Cloud Business, VIAVI Solutions
As the trends toward cloud migration and a remote workforce continue to obscure network visibility, IT issue resolution is becoming increasingly complex. How can IT teams make use of smarter analytics to speed up troubleshooting? Today’s businesses thrive on Big Data insights — from sales reports and customer demographics, to business expenses and trend analysis. This explosion of data is multiplied by escalating social media, rich multimedia files and the Internet of Things (IoT). It’s no wonder that the growth of network data traffic continues to defy expectations and IT support teams are struggling to keep up.
But beyond simply being awash in the sheer volume of data traffic, a more complex challenge for enterprise IT is how to make the best use of network performance metadata to streamline workflows, enhance productivity and meet service level expectations, while optimizing the user experience. All too often, IT support teams are drowning in a sea of data, treading water without direction while the end user waits for resolution of their issue.
To more efficiently deliver new services and turn up needed infrastructure on demand, enterprises have increasingly adopted public and private clouds, as well as software-defined networking (SDN) and network functions virtualization (NFV) technology. This approach enables enterprises to reduce total cost of ownership by only paying for the assets in use as they scale on demand to meet bandwidth needs. However, with the evolving mix of data traffic across on-premises, cloud, virtual and hybrid environments, network visibility becomes increasingly challenging.
For the personnel staffing the IT help desk, the biggest disadvantage in troubleshooting today stems from the increasing number of remote users. As more apps move to the cloud, such as customer relationship management (CRM) databases, unified communications and collaboration tools, employees have the freedom to work from virtually anywhere. How does the help desk “see” issues for a remote worker who is using the coffee shop Wi-Fi network to access an application in the cloud?
In fact, difficulty in isolating problems and determining whether they are caused by the network, system or application is a growing concern, cited as a top challenge by 65% of network and systems administrators in an annual State of the Network survey. This type of confusion extends mean-time-to-repair (MTTR), costing valuable time and money.
Helping the help desk
With a traditional approach to network management, a technician can waste a considerable amount of time trying to isolate the source of an issue. As Gartner analysts point out, network performance monitoring and diagnostics can assist with issue isolation and resolution, helping to reduce mean time to detection (MTTD). Yet IT support teams can still spend hours or even days on MTTR for just one trouble ticket as they sift through metadata and hope for the best.
A more effective strategy for network management isn’t about having more performance data but having the right data to pinpoint the root cause of the problem. For today’s high-speed, hybrid cloud networks, IT support teams require granular packet data, coupled with automated workflows that guide technicians to the answer more quickly.
As enterprises continue the migration to SDN, NFV, cloud and virtual environments, automation becomes key to maintaining network performance. This evolution path adds unprecedented dynamism to networks, allowing enterprises to reallocate capacity based on changing needs of applications and services. Just as network management needs to be responsive to these on-demand capacity changes, network performance monitoring also needs to be flexible and dynamic for improved data gathering and consumption in real-time.
Traditional network monitoring systems, which are architected to monitor static network environments, are unable to keep up with the rapidly changing conditions that require monitoring in a hybrid IT environment. Network path visualization is becoming increasingly important, particularly for those networks with a significant number of virtual elements. For example, data center virtualization technologies can overlay a second layer of infrastructure to a network.
In the new world of hybrid IT, network managers will need to monitor the performance of both the physical and virtual networks. As applications transcend physical and virtual environments, the need for path-to-service visibility as it relates to the end-user experience will become the new benchmark for any IT monitoring strategy. The ability to track these flows can be particularly challenging when they move among multiple virtual machines on a single host before exiting to a physical switch.
Tech support lifeline
As network technologies change, network performance monitoring needs to evolve in order to encompass not just physical servers and private clouds, but also all virtual, public and hybrid cloud environments. The ability to capture and store packet data across all network environments, coupled with smarter analytics to parse out relevant metrics, makes available information actionable in real-time for more accurate root cause analysis.
With smarter network monitoring, tier one tech support staff can interpret performance metadata and isolate an issue to the correct domain without escalating to the next level. This allows the trouble ticket to be assigned correctly the first time for a more efficient process workflow and overall faster MTTR.
As the predominant trend toward hybrid IT continues to cloud network visibility, issue resolution will only become more complex. For those IT teams drowning in a sea of data, smarter analytics will be a lifeline for enhanced productivity and optimized network performance.