Ask veteran SysAdmins about the good old days, and they’ll most likely talk about spending long hours in claustrophobic server rooms, healing onerous servers, and untangling network configurations.
Analysts may say we don’t remember pain, but a lifetime of troubleshooting problems under pressure—like “the internet’s broken,” slow applications, or even device issues like non-responsive printers—can take its toll.
While the days of being the underappreciated, behind-the-scenes network tamer are mostly gone, businesses are increasingly pushing their data centre digital infrastructure to cloud, begging the question: are the days of SysAdmins now numbered?
The short answer is no. SysAdmins have always adapted, moving from monitoring physical, on-premises networks to monitoring the larger data centre, where all their business servers, applications, and software stacks reside.
Unlike their network peers, SysAdmins manage higher up the stack, and generally follow their charge wherever it—and its users—wanders.
And monitor they must, as today’s cybercriminals are becoming more creative in their attacks, while data centres are often a web of vulnerabilities beyond the direct control of SysAdmins. Their Gotham City may have changed over the years, yet SysAdmins can’t afford to be any less vigilant than they were in the past.
How well they respond to new monitoring considerations, along with new tools and methodologies, will determine how effective they’ll remain in the future.
First, Know Thy Battlefield
To properly plan their monitoring approach and tools to use, SysAdmins should first consider the data centre solutions they’ve employed. Both public and private data centres will require vastly different approaches.
For starters, the lack of back-end access for public data centres (a thorn in the SysAdmin’s side) will require them to deploy detailed and real-time monitoring solutions like event log analysis to efficiently aggregate traffic and spot anomalies or suspicious activity. And if application latencies and crashes are a concern, as they always are, “transaction tracing” tools allow SysAdmins to effectively track, isolate, and remedy issues between applications, clouds, and databases.
Private clouds offer a greater degree of freedom for SysAdmins. Unfettered user access controls, along with the flexibility of the private cloud, allows SysAdmins to deploy their monitoring nets over a broader range of traffic, from hardware to applications, software, and even end-user traffic.
This allows SysAdmins to align their monitoring to business objectives: if executive concern is centered around consistent service uptimes and good customer experiences, then monitoring can be configured to watch for and respond to factors causing outages or bottlenecking systems.
More and more businesses today employ a hybrid IT approach, mixing both public and private clouds with a whole slew of systems, vastly complicating monitoring.
The right selection of tools and technologies helps, but SysAdmins may also need to change their methods and mindsets when monitoring, if they want to stay three steps ahead of their increasingly complex data centre network.
Second, Watch for the Bat Signals!
The term “monitoring” may suggest simple observation or surveillance, but data centre monitoring goes further. Especially for public data centres or clouds, the lack of control means SysAdmins have smaller timeframes to react to and rectify problems as they occur.
Many will have to adapt their thinking to include the identification of “signals,” trends or patterns signifying something is about to go wrong. As a dear colleague puts it, “Monitoring produces metrics, signals identify cause.” The job of the SysAdmin is to learn what these signals are, then move to address them before they become a supervillain of a problem.
Of course, monitoring tools and technologies help immensely in this regard. Studying historical error logs within their log analysis software, for instance, provides clues to indicate what eventually led to past hiccups, congestion, or vulnerabilities.
In the same vein, analysing real-time traffic between servers and services could also point out inefficiencies or bottlenecks within the system. Once they know what to look for, SysAdmins can feed data into automation or machine learning systems, digitally enhancing their abilities to identify and fix problems before they occur, perform even deeper analysis into the root causes of issues, and retain a confident security posture, even as the architecture of their data centres evolves and updates over time.
Form Mighty Alliances
When dealing with data centres, it’s easy for SysAdmins to see it as an “us vs. them” relationship, but it doesn’t have to be this way. Data centres are constantly seeking to optimise and secure their operations, and SysAdmins can help by sharing the monitoring metrics, insights, and signals helping keep their network ticking along.
Collaboration between SysAdmins and data centres could only result in quicker problem resolution, streamlined network traffic, and greater security for both provider and user.
Just as the job scope of the SysAdmin has changed, monitoring is now a more proactive and predictive affair, all the way from the business right up to the data centre and vice versa.
In fact, given the slew of tools and technologies available to SysAdmins today, one could even venture to say monitoring is vastly more efficient, simpler, and adaptable than physical troubleshooting and diagnoses of the past.
The question remains: can SysAdmins change and adapt, like the Dark Knight, and adopt new ways to detect, predict, and combat the issues and “bad actors” laying siege to their data centres today?
Update Your Skills With the Taxonomy
SysAdmins have always appreciated—and been frustrated by—application portability. Whether it’s delivering from custom code to bare metal servers in the past or clouds in the present, application admins always had to tinker to some extent with infrastructure, to meet end-user needs. In the process, they redefined traditional rules and reconfigured nomenclature to suit their needs.
The same applies to the relationship between today’s SysAdmins and the data centres they deal with. But even as they continue to tweak and tinker with their network layer to deliver optimal levels of service, SysAdmins must remember to bake logging and monitoring into the mix if they intend to ensure security is shored up on their end.
Doing so, along with constant adaptation and upskilling of new techniques, will provide SysAdmins with the glue to hold the “data centre” together.
