Content Copyright © 2022 Bloor. All Rights Reserved.
Also posted on: Bloor blogs
Just before the start of Covid, I wrote a White Paper, “Monitoring and managing the performance of complex hybrid IT infrastructure environments”, based on some primary research into vendor functionality. At that time, it was already becoming apparent that the rapid adoption of Cloud computing, nascent Edge computing, the proliferation of micro-services and containerisation, and the growing need for, and use of global communications networks, were creating a large hole in IT operations teams’ ability to monitor and manage the-to-end infrastructure that supported business critical applications.
Over time I had been getting more and more frustrated at the opaque nature of comments from both Network Performance Monitoring and more broadly based IT Infrastructure Performance Management vendors about what I saw as a network visibility gap in end-to-end IT infrastructure visibility.
The one exception was ThousandEyes. Here is a short excerpt from the White Paper…
For the most part, vendors with a strong network performance management heritage failed to respond to our requests to participate in this survey. We feel we should highlight and commend the participation of ThousandEyes. This vendor has a strong track record in network routing and path tracing, specifically those areas that the Hybrid Infrastructure Management (HIM) vendors don’t cover. Their capabilities include many of the areas covered by the HIM vendors, but they have not scored as highly as them in this survey, partly because they don’t yet cover the whole gamut of datacentre infrastructure and applications, but also because their AI use for analytics is less developed.”
Three years on, the market has changed dramatically. Not least, ThousandEyes itself, has been acquired by Cisco. Quite apart from the increased scale and reach provided by Cisco, integrating Thousand Eyes capabilities with those of AppDynamics helps fill in some of the gaps I mentioned in the paragraph above. In general terms, there has been an increased focus on providing IT operations departments more visibility into, and control over, the performance and availability of cloud based micro-services and containerised applications. At the same time there has been more of a focus on monitoring the digital experience of users (DEM).
But, to an extent, for enterprise network and IT operations teams, that network visibility gap has still been there. And then, the Covid-19 pandemic threw in the most enormous curveball. Overnight office and branch staff were forced to work from home. Suddenly, from having to manage network connectivity from branches and offices, which may well have been MPLS based, they now had to provide and manage connections over the public internet from employees’ homes, often using privately owned laptops, desktops and mobile devices. The lockdown saw an explosion in the use of web conferencing tools like Zoom, Microsoft Teams, Webex by Cisco and other unified comms (UC) applications that rapidly became business critical.
Most DEM solutions rely on a mix of Real User Monitoring (RUM) and synthetic monitoring, which relies on batch-based scripts, run periodically, to emulate actual users’ digital journeys. But even when DEM solutions are provided by network performance monitoring vendors, the problem of identifying exactly where in the public network the problem is occurring is onerous and time consuming. On top of that, modern UC software splits the video, voice and text chat elements and routes them to often different datacentres in order to maximise performance. This increases both mean- time to identification (MTTI) and mean-time to resolution (MTTR) of the problems at hand. This is not a good place to be for helpdesk and support teams now faced with irate senior executives who can’t get “Teams to work” from their home office.
ThousandEyes announcement this week of its Automated Session Testing solution builds on their very powerful network routing and path tracing technology. I have already written about the need to augment network monitoring with active testing. This solution from ThousandEyes goes further, in that it offers real-time visibility into every connection for every critical application session for every user with ThousandEyes Endpoint Agent running on their device, and it does this automatically on the fly.
In its initial release Automated Session Testing supports three of the most important UC solutions, Zoom, Teams and Webex. Given how critical working from home has become, and the complex nature of UC networking, this is entirely sensible. At the same time, ThousandEyes has also released Agent View, a consolidated view of every user’s digital experience. It provides a very clear visualisation of what is going on in the network as UC applications navigate the best routes to the back-end servers. In theory this will make the job of quickly identifying where problems are occurring much easier. In practice there will be even more data hitting IT and Network Operations teams already challenged by a flood of monitoring alerts.
In the next few weeks, I will be having more detailed discussions with ThousandEyes which should elicit more information about how, for example, alert prioritisation is handled, and also what other end-user applications we might see supported with Automated Session Testing in the coming months and years. In the meantime, if you are an IT or Network Operations manager dissatisfied with your ability to handle the volume of fault and support requests coming in from end users and senior executives working from home, you should probably have a look at these solutions to see if they are for you.