Cloud, AI, SDx, digitalisation: What role does IT monitoring play in keeping up with the trends?
Author : Martin Hodgson, Paessler AG
21 October 2021
Martin Hodgson, Paessler AG
If the press and analysts are to be believed, our IT landscape will look completely different in the not-too-distant future. Of course, this is also a recurring theme with regard to IT monitoring. How do we keep the balance between media hype and what works in practice? Martin Hodgson, Country Manager, Paessler AG, explains.
Cloud – still a megatrend?
Cloud is no longer a ‘trend’ as such because it has been around for a while; the cloud has received another massive boost in the wake of COVID-19 and the resulting home office. I don’t think there has been a fundamental paradigm shift due to the pandemic: the cloud was already dominating the headlines before 2020 and has merely moved into focus a bit more due to COVID-19.
Whilst the uptake of the cloud has been widespread, local IT still exists and hasn’t lost any of its importance – even analysts and journalists who were still proclaiming the end of the admin just a few years ago have now come around and are talking about hybrid environments, i.e. the combination of public cloud services, private cloud and local IT.
This is also what we are observing – with our customers and also with our own IT. For example, services are sourced from the cloud: Microsoft 365, CRM system, store systems, etc., and data are stored in the public cloud.
The private cloud plays an important role for many companies when it comes to storing sensitive data or services that should/should not be outsourced to the public cloud.
Finally, for most companies, local IT also plays a far greater role than just providing Internet access. Sensitive data is stored locally, applications run in the company's own data centre and, of course, there is also the local infrastructure.
Monitoring is required to support IT Ops: of course, local IT must not be neglected, but that is usually the bread & butter business of most established monitoring solutions anyway. In addition, cloud applications and cloud providers must be included in the central monitoring by querying the corresponding interfaces. This means that the monitoring solutions must be continuously expanded and adapted.
AI: over-hyped or one to watch?
AI has become more and more present in recent years, thanks partly due to the development of cloud technology. However, "real" AI in the form of machine learning is often still lumped together with more or less complex algorithms. Whilst there are endless opportunities and possibilities here, there is also a lot of hype and misconceptions.
AI can play an important role in network monitoring, as it collects gigantic amounts of data, and the combination of cloud and AI or machine learning is predestined to analyse this data and recognise patterns. This can be about detecting anomalies, improved root-cause analysis, but also trending and predictive maintenance.
At present, it is mainly a few new and cloud-based monitoring solutions, primarily for the enterprise market, that are on the move in terms of AI. These are mostly security or application performance monitoring based on advanced traffic monitoring, i.e. highly specialised solutions aimed at proven specialists.
SDx: making IT monitoring easier
What started as software-defined networking has now evolved into SDx, software-defined everything – storage, WAN, LAN, radio, data centres, etc. While, in the past, devices became more and more intelligent and therefore more complex, intelligence is increasingly outsourced to include a software layer. The advantages are obvious: "dumber" and therefore cheaper devices are centrally set up and controlled. There are lower acquisition costs for hardware, less effort for configuration and management.
While, in the past, more and more intelligent devices had to be integrated into the monitoring system individually and at great expense, this has changed somewhat with SDx: by connecting the software layer, many values can be queried directly there. In parallel, additional factors such as underlying hardware, traffic, and the surrounding infrastructure can be monitored via a broad-based monitoring solution, thus providing a comprehensive picture of the entire SDx environment. In the end, not much has changed for broad-based monitoring solutions with SDx. There is a new component to IT that usually brings its monitoring capabilities, which in turn can be integrated into an overarching monitoring effort.
Digitalisation is here to stay
Digitisation is often seen only as a marginal phenomenon, since it involves areas that were previously separate from IT – e.g., production facilities, medical devices, infrastructures, or building technology. However, as digitisation generates more and more data its ability to transport, store and process data lies with IT. In other words, sooner or later digitisation will have a significant impact on IT.
Of the four trends listed, digitisation is least associated with traditional IT, yet this is actually where it has the strongest influence. In terms of monitoring, in the event of a malfunction, it is possible to immediately identify where the problem is and whether there are bottlenecks in data transport in the network. However, this only works in a hospital context if the monitoring solution can monitor both IT and medical infrastructures – but there are no comprehensive monitoring systems in the medical sector. The communication servers commonly used there have no or only very rudimentary monitoring functionality and certainly no ambition to extend this to heterogeneous and highly complex IT environments. Therefore, overarching monitoring solutions must come from IT.
In the future, SDx, cloud and digitalisation will be standard
Monitoring solutions with a broad approach will cover everything from IT to production to smart building technology or medical technology. It will provide a central overview as well as dashboards for experts. In the long term, the topic of AI will also gain importance for broad-based solutions. Monitoring produces gigantic amounts of data. This data currently provides information on the current status of the monitored components. With the help of artificial intelligence, it will also be possible to predict trends as they emerge in areas such as predictive maintenance, best practice, unusual behaviour and intrusion detection.
Monitoring is fundamentally a technology follower, not a driver
There is always a certain need for innovative monitoring tools that serve cutting-edge technologies. However, the majority of IT administrators and IT Ops teams live in the here and now and have to face the challenges of the present. Broad-based monitoring solutions like PRTG must first and foremost meet these challenges and evolve to meet the needs of their customers. This currently means comprehensive features for monitoring classic (and hybrid) IT environments, integration of cloud services and SDx providers as well as increasing integration of protocols for digitalisation.
Contact Details and Archive...