By Len Rosenthal, CMO
This blog originally published on APMdigest – read the original here.
Most enterprises lack the complete visibility required to avoid business-impacting application outages and slowdowns – resulting in nearly 90 percent of enterprises being unable to consistently meet service level agreements (SLAs) for their business-critical applications, according to a recent survey conducted by Dimensional Research and Virtual Instruments. This research indicates a serious gap in IT operations teams’ ability to monitor their enterprises’ highly virtualized, multi-vendor hybrid data center environments, and the results show that this lack of visibility is significantly impacting business.
Blind Spots, Slowdowns and Outages Abound
The reality is that large enterprises endure a substantial number of application outages and performance issues every year, and an overwhelming number of those surveyed indicated that a slowdown impacts businesses just as much as a full outage.
86 percent of users experience two or more significant outages a year, with 61 percent suffering from four or more in the same period.
59 percent of application outages and performance problems are related to infrastructure, which begs the question: why can’t IT teams see these problems coming, and what’s getting in the way of timely resolution?
Too Many Cooks in the Kitchen
There are many dozens of infrastructure and application monitoring tools available to enterprises, so why does this visibility gap still exist?
This research showed that it’s not necessarily a lack of tools that may be causing the problem, but rather the combination of too many silo-specific tools. In fact, more than 70 percent of respondents use more than five IT infrastructure monitoring tools, and 15 percent use more than 20!
But despite this plethora of tools, 54 percent of companies lack full visibility into their infrastructure and application workload behavior, and 42 percent of companies operate primarily in “reactive mode” when managing their infrastructure.
Teamwork Makes the Dream Work
When it comes to the modern enterprise, there’s no single internal team that can accurately manage and assess application performance requirements. However, less than half of enterprises take a collaborative approach to establishing performance requirements for new data center infrastructure. With no collective understanding of how applications relate to the underlying infrastructure, the resulting blind spots cause chain reactions that leave enterprises highly exposed.
Nearly 40 percent of enterprises say that performance issues related to infrastructure are the most challenging to resolve, and when you consider that 79 percent of application outages and other issues directly impact customers, there just isn’t room for guessing.
Deeper Insights Are the Key
The lack of visibility and proactive infrastructure and application management contributes to a lack of confidence from IT teams and their executives. In fact, 62 percent doubt that their current infrastructure would be able to meet their projected performance needs in the next two years, and two-thirds of respondents feel that they’re often held personally responsible for application outages and slowdowns.
In addition, with an increasing number of applications being deployed in public clouds, nearly 65 percent are concerned about the perceived value of the internal IT infrastructure team to the business.
As discouraging as these findings may seem, the numbers indicate a strong opportunity for engineering, operations and application teams to come together and gain a deeper understanding of the impact of their applications on the underlying infrastructure, and visa versa. Since applications and infrastructure are intertwined to the point where they can no longer be viewed as distinct entities, an infrastructure monitoring approach that understands application workload behavior is essential to performance assurance.
The bottom line is that in today’s highly competitive business environment, enterprises cannot afford to test their customers’ limited patience by having an unacceptable number of application outages or slowdowns.