I started this blog and while back and never finished. Here is a quick brain dump of some of my thoughts on the state of monitoring and visibility in data networks.
I feel that network operators from time to time forget the purpose of their role. It’s understandable in the day to day grind. We keep building our network up to provide better capacity and up time but in that never ending cycle trying to keep up we might not think so much about the end result – the user experience.
To that end, network performance and management tools are focused on helping us to make sure that we keep up with the status quo of keeping the packets flowing and staying ahead of capacity needs. However, as IT systems in general become more and more complex due to mobility of endpoints and workload, increasingly complex and integrated systems and security concerns, status quo is likely no longer good enough.
We need to understand at a more intricate level how the network may or may not be impacting application performance. An obvious example of the changes that are driving the need for greater visibility is an ever increasing presence SaaS offerings that are critical to business operation.
It has become very obvious to me that we need tools that no longer just measure network performance on capacity and up time. We desperately need tools quickly and easily identify network performance from the view of the endpoint and the application.
Enter the cloud! This is where the story gets exciting with several very helpful SaaS solutions now available that are capable of vastly enhancing an organization’s visibility and many only take minutes, hours or days to enable and begin getting telemetry data.
App Dynamics is a client based monitoring platform for applications. I have seen remarkable improvements in application up-time metrics in organizations that have adopted this platform. App Dynamics excels in passive monitoring of application performance and works well at analyzing baseline application performance and creating KPI metrics that can be used to tune standard deviation alerting.
Thousand Eyes is a network monitoring and synthetic traffic generation tool. Unlike App Dynamics it is technically client-less (meaning it doesn’t need to be installed or integrated with on servers or network devices). Rather at it’s core Thousand Eyes VMs send synthetic traffic to a target IP or URL and records a large amount of data about the target. It maps the network paths, hops and links, reports on latency metrics from the VMs perspective. It is very useful in monitoring critical applications up time and performance and giving insight into networks that are not under a common administrative domain (IE provider or multi-tenant networks).
App Dynamics and Thousand Eyes are both branching out to enhance their capabilities. App Dynamics for example is adding network transaction detail and Thousand Eyes has both a user agent that can passively capture transaction performance from and endpoints perspective as well as can be used for limited synthetic traffic generation. Thousand Eyes is also working on a ‘device layer’ capabilities, in short the ability to pull down SNMP information from devices under the same administrative domain to give further insight into device and link performance.
Of course there are many other solutions on the market. I just wanted to highlight two that I have seen in action and both of which have provided significant value and impact almost immediately.
The industry still seems to have much work to do in this area and I think that this will be an ongoing topic for years to come. But the accelerated pace at which innovation can happen in SaaS offerings is exciting to witness.