During the next two years, global spending on cognitive systems tools, such as data analytics and machine learning, is set to reach$31.3 billion. Today, tools based on these technologies are popping up in every corner of businesses. We’ve seen them emerging in many areas – from artificial intelligence (AI)-powered customer service bots to trucking routes designed by data scientists. Ironically, there’s one department that has yet to fully leverage the power of machine-learning based analytics — IT.

However, that is changing as IT environments are becoming increasingly complex, they’re moving beyond physical servers to virtual environments. According to a recentstudy from SIOS Technology Corp., 81 percent of IT teams are now running their business-critical applications in VMware environments.

Unlike physical environments, virtual environments are comprised of VMs, applications, storage and network that are highly interrelated and constantly changing. To manage and optimize these environments, IT managers have to analyze an enormous volume of data and understand the patterns of behavior within IT environments. As a result a new field has emerged -- AIOps.

What is AIOps?

AIOps (algorithmic IT operations platforms) is a new term coined byGartner to describe the next phase of IT operations analytics that leverages machine learning technology to automate intelligence gathering and problem resolution for IT operations.

Right now, Gartner estimates only five percent of businesses have an AIOps platform in place. However, during the next two years that number is expected to mushroom to 25 percent. AIOps replaces human intelligence with machine intelligence to decipher the complex interactions within a business’ virtual environment to instantaneously uncover infrastructure issues, correlate them to problems in application operations and recommend solutions.

Using machine learning, AIOps platforms learn how these environments behave over time to identify abnormal behavior and understand what is causing the issues within their environment. Advanced AIOps platforms can even find and stop potential threats before they become application performance issues.

 

Why do businesses need AIOps?

As organizations move their business-critical applications into virtualized environments, finding the root cause of application performance issues is more complicated than ever before. Today, IT has become a complex web of VM applications, storage devices, network devices and services that are connected in ways IT can’t always understand.

Often, the relationships within a VMware or other virtual environment becomes so intertwined that moving a workload or making a change to one object may affect objects in several other silos without their knowledge. This makes the process of tracing back  IT performance issues to its root cause difficult, if not impossible for IT leaders.

According to the recent SIOS report, 78 percent of IT professionals are using multiple tools—including application monitoring, reporting and infrastructure analytics, to identify the cause of application performance issues in VMware.

Often, when faced with an issue, IT assembles a team with representatives from each IT silo who use their own diagnostic tools and bring their own perspectives to the table. The team compares the results of these tools to find common elements, timestamps and changes in infrastructure that may indicate the root cause of performance issues. The specialized skills required, manual inefficiencies and poor responsiveness of the “war room” approach is driving IT’s interest in using machine learning tools to automate problem solving and save time, resources and manpower in IT.

Application performance issues are wasting time and resources for organizations and imposing significant challenges to IT, which is why businesses are increasingly turning toward AIOps, using machine learning-based analytics tools, to solve these issues. This marks an important shift from looking at discrete objects and events after the fact to seeing across the infrastructure in real time and automatically identifying behaviors that may indicate a problem – and recommending specific steps for resolution.

What’s Next for AIOps?

Every aspect of business is creating unprecedented volumes of data and level of complexity that cannot be managed effectively with traditional, manual methods; and IT is no exception. For IT, having tools that analyze infrastructure and application data in real time and automatically provide recommendations for solving problems and improving efficiency marks an important shift. 

Now and in the next few years, the IT profession will rapidly move from a 20+ year old manually intensive “computer science” approach using limited monitoring based tools to a modern “data science” approach using advanced analytics driven by machine learning. For IT teams, this means embracing machine learning-based analytics solutions, and understanding how to use it to solve problems efficiently and effectively. Looking to the year ahead, executives need to work with their IT departments to identify to right AIOps platform for their business.

(About the author:Sergey Razin is the CTO ofSIOS Technology Corp. and is a pioneer in the application of advanced analytics and machine learning in the areas of IT security, media and speech recognition. Prior to joining SIOS, Sergey was an architect for EMC storage products and EMC CTO office where he drove initiatives in areas of network protocols, cloud and storage management, metrics and analytics.)