This article was originally published on VMblog.com on August 14, 2018.
Growing numbers of enterprises are adopting SD-WAN technology, seeing it as a key to increasing business agility according to Dr. Jim Metzler, co-founder and principal analyst with consultancy Ashton, Metzler & Associates. However, monitoring and troubleshooting are one of the issues threatening to stall the expansion of SD-WAN use-at least, for those who have tried integration on their own.
Metzler’s research, highlighted in his 2018 State of the WAN report, found that there are still cases where enterprises are not aggressively implementing SD-WAN. In some cases, enterprises were having challenges with implementing SD-WAN on their own and integrating the solution with their existing WAN, with some typical issues including:
- Challenges in troubleshooting performance/configuration issues
- Unanticipated difficulties in installation
- Increased complexity of operations
SD-WAN gives enterprises new options for network services, including blending of broadband, Wi-Fi, and cellular links from different providers. However, monitoring and maintaining application performance across heterogeneous network and cloud systems creates new challenges for IT operations personnel.
The challenge of network monitoring in the cloud era
There are tools for APM (application performance management) and NPM (network performance management), but these tools aren’t always going to help trace the root cause of every service interruption because they (generally speaking) only provide analysis of their respective technology domains. These separate tools have separate historical data sets, making it difficult to gain insight into interactions between network and application that can result in performance degradation.
There’s another factor that can add to the challenge of monitoring performance: whether or not the enterprise is using an ‘on-network’ SD-WAN service (provided by an asset-based network service provider) or an overlay SD-WAN service provider, which does the work of sourcing and maintaining links to points of presence (POPs) around the globe. While both can offer assistance in the form of 24/7 operations/monitoring on a global basis, there are still questions to ask a service provider.
Questions to ask vendors include:
- Can the on-net provider do anything to monitor and remedy off-net issues?
- A service from an overlay provider can provide greater control and monitoring than an on-net provider, but can you integrate in to other monitoring systems?
Customers need a network telemetry and monitoring agent that can scan and report on both network and application performance issues across underlay and overlay networks. A fundamental element required to take full advantage of telemetry: open APIs that can take export data from the SD-WAN orchestration platform and integrate with other datasets from switches, routers, firewalls, and other network and application management platforms.
Of course, just having more data doesn’t help by itself. The next element of an integrated network and application monitoring system iterates the multiple data sets analytics platform for visualization and to find complex correlations.
The combination of an SD-WAN-based network infrastructure, and integration of multiple sources of network and application performance data into an analytics platform will help infrastructure and operations (I&O) teams find performance problems faster. And that further extends the transformative promise of SD-WAN as a powerful tool fueling digital transformation.
To learn more about Apcela’s monitoring tools, click here.
