Using Machine Learning Analytics to Deliver Service Levels
September 21, 2016

Jerry Melnick
SIOS Technology

Share this

While the layers of abstraction created in virtualized environments afford numerous advantages, they can also obscure how the virtual resources are best allocated and how physical resources are performing. This can make maintaining optimal application performance a never-ending exercise in trial-and-error.

This post highlights some of the challenges encountered when using traditional monitoring and analytics tools, and describes how machine learning, as a next-generation analytics platform, provides a better way to meet SLAs by finding and fixing issues before they become performance problems. A future post will describe how machine learning analytics can also be used to allocate resources for optimal performance and cost-saving efficiency.

Most IT departments identify performance problems with tools that monitor a variety of discrete events against preset thresholds. For example they set a specific threshold for CPU utilization. Whenever that threshold is exceeded, the tool fires off alerts. But the use of thresholds presents several challenges. They do not account for the interrelated nature of resources in virtualized environments, where a change to or in one can have a significant impact on another. Such interrelationships exist both within and across silos. Without a complete understanding of the environment across silos, users of threshold-based tools frequently discover that their attempts to solve a problem have simply moved it to a different silo.

Thresholds often generate "alert storms" of meaningless data and miss important correlations that might indicate a severe problem exists. They are ineffective in detecting the symptoms of subtle issues that may indicate a significant imminent problem such as "noisy neighbors" or datastore latency issues. These subtle issues may not exceed a threshold related to the root cause or may exceed a threshold in short, random intervals, producing alerts that are frequently lost amid the "noise" of alert storms.

Even the so-called dynamic thresholds cannot accommodate the constant change in dynamic environments and, as a result, require significant ongoing IT intervention. And finally, while they may alert IT to an issue, they rarely provide sufficiently actionable information for resolving it. The exponential growth in the size and complexity of virtual environments has outstripped the ability of IT staff to set, manage, and continuously adjust threshold-based tools effectively. The time for an automated solution has come.

Advanced machine learning-based analytics software overcomes these and other challenges by continuously learning the many complex behaviors and interactions among interrelated objects – CPU, storage, network, applications – across the infrastructure. Unlike threshold-based solutions, this growing knowledge enables machine learning-based IT analytics solutions to provide a highly accurate means of identifying the root cause(s) of performance problems and making specific recommendations for resolving them cost-effectively.

This ability to aggregate, normalize, and then correlate and analyze hundreds of thousands of data points from different monitoring and management systems enable machine learning analytics solutions to transform massive volumes of data into meaningful insights across applications, servers and hosts, and storage and network infrastructures.

As it gathers and analyzes this wealth of data, the MLA system learns what constitutes normal behaviors, and it is this baseline that gives the system the ability to detect anomalies and find root causes automatically.

In addition to identifying root causes, advance machine learning based analytics solutions are able to simulate and predict the impact of making certain changes in resources and their allocations, which can be particularly useful for optimizing resource utilization and planning for expansion. This capability can also be useful for assessing if there is adequate capacity to handle a partial or complete failover. And these are topics worthy of a deeper dive in a future post.

Jerry Melnick is President and CEO of SIOS Technology.

Jerry Melnick is President and CEO of SIOS Technology
Share this

The Latest

October 17, 2019

As the data generated by organizations grows, APM tools are now required to do a lot more than basic monitoring of metrics. Modern data is often raw and unstructured and requires more advanced methods of analysis. The tools must help dig deep into this data for both forensic analysis and predictive analysis. To extract more accurate and cheaper insights, modern APM tools use Big Data techniques to store, access, and analyze the multi-dimensional data ...

October 16, 2019

Modern enterprises are generating data at an unprecedented rate but aren't taking advantage of all the data available to them in order to drive real-time, actionable insights. According to a recent study commissioned by Actian, more than half of enterprises today are unable to efficiently manage nor effectively use data to drive decision-making ...

October 15, 2019

According to a study by Forrester Research, an enhanced UX design can increase the conversion rate by 400%. If UX has become the ultimate arbiter in determining the success or failure of a product or service, let us first understand what UX is all about ...

October 10, 2019

The requirements of an APM tool are now much more complex than they've ever been. Not only do they need to trace a user transaction across numerous microservices on the same system, but they also need to happen pretty fast ...

October 09, 2019

Performance monitoring is an old problem. As technology has advanced, we've had to evolve how we monitor applications. Initially, performance monitoring largely involved sending ICMP messages to start troubleshooting a down or slow application. Applications have gotten much more complex, so this is no longer enough. Now we need to know not just whether an application is broken, but why it broke. So APM has had to evolve over the years for us to get there. But how did this evolution take place, and what happens next? Let's find out ...

October 08, 2019

There are some IT organizations that are using DevOps methodology but are wary of getting bogged down in ITSM procedures. But without at least some ITSM controls in place, organizations lose their focus on systematic customer engagement, making it harder for them to scale ...

October 07, 2019
OK, I admit it. "Service modeling" is an awkward term, especially when you're trying to frame three rather controversial acronyms in the same overall place: CMDB, CMS and DDM. Nevertheless, that's exactly what we did in EMA's most recent research: <span style="font-style: italic;">Service Modeling in the Age of Cloud and Containers</span>. The goal was to establish a more holistic context for looking at the synergies and differences across all these areas ...
October 03, 2019

If you have deployed a Java application in production, you've probably encountered a situation where the application suddenly starts to take up a large amount of CPU. When this happens, application response becomes sluggish and users begin to complain about slow response. Often the solution to this problem is to restart the application and, lo and behold, the problem goes away — only to reappear a few days later. A key question then is: how to troubleshoot high CPU usage of a Java application? ...

October 02, 2019

Operations are no longer tethered tightly to a main office, as the headquarters-centric model has been retired in favor of a more decentralized enterprise structure. Rather than focus the business around a single location, enterprises are now comprised of a web of remote offices and individuals, where network connectivity has broken down the geographic barriers that in the past limited the availability of talent and resources. Key to the success of the decentralized enterprise model is a new generation of collaboration and communication tools ...

October 01, 2019

To better understand the AI maturity of businesses, Dotscience conducted a survey of 500 industry professionals. Research findings indicate that although enterprises are dedicating significant time and resources towards their AI deployments, many data science and ML teams don't have the adequate tools needed to properly collaborate on, build and deploy AI models efficiently ...