IT Briefcase Exclusive Interview: The Rise of Machine Learning Analytics

Today’s IT infrastructures are more complex than ever. Companies are looking for new ways to combine a wide variety of resources spanning physical, virtual, and cloud platforms to support their IT operations. These new data centers reduce costs and increase mobility. IT departments can quickly align operations, deploy applications, and configure systems in a way that best meets constantly changing business needs. But with this convenience and flexibility, new challenges arise.

In this interview, SIOS Technology President and CEO Jerry Melnick speaks with IT Briefcase about optimizing infrastructure operations and application performance in complex VMware environments. These environments are dynamically changing and objects within them (VMs, hosts, network, storage, etc.) are highly interdependent making it difficult to find and fix issues impacting application service delivery. Jerry takes a closer look at the current approaches and new machine learning-based approaches that are bringing predictability, order and intelligence to IT operations management.

• Q. How has managing the complexity of application environments been addressed traditionally?

A. Optimizing application environments in VMware has been a highly manual, time-consuming process, requiring significant guesswork and continuous adjustment. Traditional operational analytics approaches focus on recording and reporting discrete events – for example, CPU utilization exceeding a particular threshold. These tools miss the complex interrelationships in the infrastructure requiring highly skilled IT personnel long hours to uncover the issues impacting critical application operations. Legacy approaches fall short of the needs of IT managers running business critical applications in complex, dynamic virtualized environments. They also require the collaboration of highly skilled cross-functional teams of domain experts from each of the computing tiers; application, storage, network, and virtualization infrastructure.

• Q. Why are some applications more challenging to manage and optimize in virtual environments than others?

A. As enterprise IT environments look to move more of their business operations to virtual environments, many struggle to meet the stringent requirements of business-critical applications such as SQL Server, Oracle, and SAP in these environments. Because they are essential to the company’s productivity, IT managers are required to ensure these applications and the environments they operate in are optimized for high performance, reliability, and efficiency.

However, meeting these requirements in a virtual server environment is much more difficult than in a physical server environment where applications run on dedicated resources in tightly controlled and managed infrastructures Virtual environments are highly scalable with shared and dynamic resources that operate in complex relationships with one another. The operation of one can affect the resources used by one or more of the others. Understanding how these subtle relationships affect performance and the ways to correct performance issues can be extraordinarily difficult.

IT is also challenged with planning and scaling fast-growing, dynamic virtual infrastructures while supporting new applications and maintaining reliability and availability. Today, IT often uses estimates and trial-and-error methods to identify problems, improve configurations, and evaluate the impact of changes.

• Q. What is Machine Learning-based analytics?

A. Machine learning-based analytics makes a vast improvement on the one-dimensional threshold based tools by focusing on knowledge discovery rather than reporting merely report data or metrics. The next-generation machine learning analytics tools deliver actionable information that IT administrators can use to solve problems. In doing so, they eliminate hours of manual time IT spends comparing reports and information from multiple sources to identify issues.

• Q. What are some of the benefits of Machine Learning analytics approaches?

A. Machine Learning technology is a fundamentally different approach to IT analytics. It is an adaptive technology that “learns” about the infrastructure – and the interrelationships of its various components – over time. Rather than simply reporting discrete events using a point-in-time status or data averages, it provides knowledge of the infrastructure that is extremely valuable and powerful in helping predict, simulate, and recommend solutions.

• Q. How does implementing traditional approaches compare to new Machine Learning-based approaches?

A. Implementing existing solutions can take hours of manual threshold setting; installation of agents; and configuration of traps, events, and logs. Establishing thresholds for these tools involves a best-guess and a trial-and-error approach to approximating “normal” operations. In many tools, users are left to the time-consuming task of building and customizing dashboards. Once configured, they can be extremely complex to use and require extensive training.

The more advanced machine learning-based analytics solutions are highly automated and completely eliminate the manual configuration and rules definition required by conventional monitoring tools. Because they require no agents or complex configuration, IT can set up some machine learning technology in as little as 15 minutes and start gaining important insights about infrastructure and application operations without requiring time-consuming manual analysis from multiple domain experts.

• Q. Tell us about your solution, SIOS iQ.

A. SIOS iQ is the only product to automatically diagnose and recommend solutions to application performance issues and model the outcomes of recommended changes in virtualized environments.

Designed to be a powerful platform for IT operations information and issue resolution, SIOS iQ applies an advanced data analytics/Big Data approach to a broad range of data sets, including application and infrastructure data from third party tools and frameworks, to recognize abnormal patterns of behavior and identify root causes of performance issues. It provides information organized according to four key dimensions: performance, efficiency, reliability, and capacity utilization. The new release integrates SIOS iQ with SQL Sentry Performance Advisor, bridging a critical gap between IT infrastructure administrators and SQL Server administrators. For the first time, IT staff can instantaneously identify and resolve the root causes of performance issues based on a comprehensive analysis of both the VMware infrastructure and the SQL Server application environment.

• Q. Can you tell us about customers who are benefitting from machine learning-based analytics?

A. Global security software company, Trend Micro; Datrix, one of the UK’s leading providers of critical network services and enterprise cloud solutions; and MA-based Jordan’s Furniture, are among those gaining faster, more accurate insights into their VMware environments. Trend Micro has to meet stringent Service Level agreements for application performance and efficiency both for its employees and its customers. To find the root causes of performance issues, they had used the traditional ‘war room’ approach of assembling a multi-disciplinary team of IT staff and analyzing multiple analytics tools. With machine learning-based analytics Trend Micro can now find and resolve root causes of performance issues in minutes, saving hours of IT labor and enabling them to maintain compliance with their SLAs. Jordan’s Furniture is using advanced machine learning analytics in its VMware vSphere environment and within minutes, received recommended changes that addressed the root cause of performance issues resulting in higher performance gains without the need to purchase additional hardware. Jordan’s Furniture was able to improve performance without wasting time and money.

Jerry Melnick
President and CEO, SIOS Technology Corp.

Jerry is responsible directing the overall corporate strategy for SIOS Technology Corp. and leading the company’s ongoing growth and expansion. He has more than 25 years of experience in the enterprise and high availability software markets. Before joining SIOS, he was CTO at Marathon Technologies where he led business and product strategy for the company’s fault tolerant solutions. His experience also includes executive positions at PPGx, Inc. and Belmont Research, where he was responsible for building a leading-edge software product and consulting business focused on supplying data warehouse and analytical tools. Jerry began his career at Digital Equipment Corporation where he led an entrepreneurial business unit that delivered highly scalable, mission critical database platforms to support enterprise-computing environments in the medical, financial and telecommunication markets. He holds a Bachelor of Science degree from Beloit College with graduate work in Computer Engineering and Computer Science at Boston University.