IT organizations still struggle to keep the lights on. While the business expects IT to enable digital transformation, the reality is that IT spends 70% to 80% of its time keeping services up and running. That’s a huge drag on innovation and productivity, and as services become more complex and distributed, the challenge is increasing.
However, ask yourself this. What if IT could step back entirely, leaving networks and data centers to look after themselves? Instead of being consumed with labor-intensive operational activities, IT is unleashed, transforming its role and business value.
Are We Ready to Turn the Lights out?
In the past, this vision was known as autonomic computing – creating a data center where applications and machines automatically adapt to environmental changes. Human operators define policies and rules, guiding machine behavior. Machines follow these policies, automatically doing all the grunt work – including configuration, optimization, and protecting against external attacks. Perhaps most important, machines automatically manage service availability and performance, automatically restoring service when there’s a failure or degradation.
Autonomic computing is still a vision, but we’re making huge strides in service availability and performance. This is often referred to as software-defined IT operations, and it’s starting to transform how we manage data centers and networks. By applying artificial intelligence and machine learning to real-time and historical operational data, software-defined IT operations is helping IT teams to reduce downtime and IT spend, using AI to detect, diagnose, remediate, predict and eliminate outages.
Why Is This Important?
It’s because IT is being asked to launch new applications and business services at an ever-increasing rate. And, as we move to cloud and hybrid IT environments, applications and services are becoming more distributed and complex, resulting in a dramatically increased flood of monitoring data. There’s no way that humans are going to be able to keep up. While software-defined data centers and software defined networking are dramatically increasing agility, they are also breaking traditional manual IT operations models.
At Evanios, we embrace software-defined IT operations. In fact, the Evanios platform delivers some of the most advanced software-defined IT operations capabilities on the market today. Here are some of the key capabilities we deliver, and how our customers are benefiting from these:
Automated noise reduction – While next-generation cloud applications and services generate a deluge of events, Evanios reduces these to a trickle. We use a combination of preconfigured logic, machine learning, and other intelligence noise reduction techniques to normalize, filter, deduplicate and correlate events, creating a clean signal. In some cases, the noise reduction can be as high as 100,000 to 1. Rather than spending enormous amounts of time trying to make sense of millions of disconnected events, IT operations teams focus on actionable alerts – so they can scale to handle today’s rapidly accelerating monitoring data volumes.
Intelligent (automated) event prioritization – Evanios doesn’t just dramatically reduce events, it intelligently ranks them. While monitoring systems do prioritize events, they lack any business or historical context. They don’t know how events affect particular business services, or how important those services are. And, without any historical insights, they don’t know how likely an event is to cause a service outage or degradation. In contrast, Evanios uses machine intelligence to learn from configuration and historical data, identifying key event traits – such as affected business services and the priority of similar historical incidents – that predict the likely business impact of each actionable event. That means IT operations teams get a clear ranked list of what’s important, so that they focus on what matters.
Automated diagnosis – IT spends huge amounts of time trying to identify the root cause of service issues. That leads to significant delays and lengthy service outages. Again, today’s increasingly complex hybrid IT infrastructures are making this problem worse. Evanios leverages artificial intelligence to automatically identify the likely reasons of incidents, tracing symptoms to root causes using a combination of real-time analysis and historical comparison – so that issues are diagnosed and remediated faster.
Prediction for prevention – This is a truly unique capability that only machine learning can deliver. By leveraging the power of AI to identify patterns in large data sets, Evanios identifies the precursors to service issues, raising predictive events that pinpoint the likelihood, severity, and root cause of future service outages and degradations. By predicting these issues, Evanios gives IT teams the opportunity to prevent them, increasing service quality and preventing serious business impacts.
Assisted remediation – Finally, Evanios learns from how similar incidents were remediated in the past and uses this to provide remediation recommendations. It can also trigger automated remediation, either autonomously or after approval by IT operations staff. Again, this dramatically reduces MTTR and eliminates human error.
Taken together, these intelligent capabilities automate the IT service assurance lifecycle, which is the core goal of software-defined IT operations. They assist IT operations staff at every step, dramatically accelerating issue resolution. And, as organizations build confidence in these automated capabilities, IT staff can take an increasingly hands-off approach, and instead focus on creating policies and validating outcomes. At this point, the promise of software defined IT operations becomes a reality, laying a cornerstone for broader autonomic computing.