Tablet Customer Satisfaction

Trends in Application and Infrastructure Performance Monitoring

Posted on 08. Apr, 2018 by in big data, Cloud, Developement, IAAS, internet, linux, mobile, Open stack, Programming, SAAS, Uncategorized

Trends in Application and
Infrastructure Performance Monitoring


Big data and related technologies, such as machine learning and SaaS delivery models, are front and center in many discussions about application and infrastructure performance management today.

The prevalence of complex application environments that feature containers, microservices and multiple clouds is driving these discussions as enterprises struggle with the data overload that such environments produce.

Performance monitoring tools are adapting, harnessing new technologies that can better collect, store and analyze the deluge of data.

This report focuses on the following segments of the application and infrastructure performance monitoring market:

  • Code-level application performance monitoring (APM)
  • Real user monitoring (RUM)
  • Synthetic monitoring
  • Performance testing
  • Network performance monitoring (NPM)

Code-Level APM

APM products reveal metrics about application performance, including load, response time and errors. Code-level APM identifies the individual lines of code in applications that may be causing problems.

New technologies such as containers and microservices are shaping the APM market, and some of these innovations require new approaches to APM.

For example, microservices users demand an understanding of the relationships between services in order to better pinpoint the root cause of problems. That demand is driving interest in transaction tracing, which APM vendors are integrating.

Technologies such as containers and microservices also increase the complexity of applications, and in turn enlarge the volume of operations data that IT organizations need to collect in order to track application performance.

Vendors that offer APM – and other services in the application and infrastructure performance market – are increasingly embracing machine learning techniques to rapidly mine data for useful information.

Real User Monitoring (RUM)

RUM shows metrics, such as load time, that actual users are experiencing. It displays visualizations of the user experience – based on geography, browser, operating system and device type – in order to help isolate problems.

RUM is often used in conjunction with other tools, such as synthetic monitoring, to validate results from the synthetic traffic and APM tools to allow developers to determine the impact of a performance problem and to prioritize repairs.

Integration across RUM and adjacent functions is becoming more important as application developers show growing interest in the user experience.

Synthetic Monitoring

Synthetic monitoring uses scripts to simulate user activity on a site or application, delivering results about performance from the browser or application perspective.

Sending synthetic traffic to applications allows developer teams to track the performance of an application and identify whether that performance differs based on the browser, device or location from which the synthetic traffic originates.

Businesses use synthetic monitoring to catch performance issues before they affect real customers (e.g., testing how an application or update might perform in targeted geographic areas).

IT organizations often use synthetic monitoring to establish a performance baseline, as well as to combine the results with data from APM tools.

This approach allows application development teams to compare performance from the end users’ perspective via the synthetic traffic, with information about the performance of back-end systems collected by APM tools.

Synthetic monitoring is also commonly used in combination with RUM tools, where developers detect problems via the synthetic traffic and can then check whether those problems are affecting real users.

Additionally, information gathered via RUM tools can improve the use of synthetic tools by ensuring that application development teams are designing synthetic traffic that reflects real use cases.

Performance Testing

Performance testing generates large volumes of traffic to ensure that an application will perform as expected, even under heavy loads.

Performance testing is increasingly in the spotlight as M&A activity ramps up. Recent deals included Akamai’s acquisition of SOASTA and Tricentis’ acquisition of Flood.io.

This activity is fueled by demand for quicker and easier testing that won’t become a bottleneck in software release cycles.

Some IT organizations report a drop in application performance that corresponds with the adoption of practices – such as DevOps – designed to speed up software development and release.

Some companies may be overlooking proper testing in order to speed up the development process. The need to address this issue is driving new capabilities in performance testing tools, including integration with CI/CD tools to automate testing.

This integration supports continuous testing and ‘shift left,’ where testing starts earlier in the development process and is executed throughout, with the intent of repairing problems before they turn into larger issues.

Network Performance Monitoring (NPM)

NPM tools use a range of approaches – including collecting and analyzing flow and SNMP data, generating synthetic traffic and inspecting packets – to deliver insight into bandwidth utilization, bottlenecks and issues affecting network components.

The NPM sector also includes more sophisticated products, including physical taps or other appliances that enable deeper visibility into networks.

The emergence of this variety of approaches is partly due to the ongoing shift to the cloud, which in turn shifts an increasing volume of enterprise traffic to public networks.

Enterprises therefore require tools that can offer insight into external networks as well as their own networks.

Vendors push network monitoring and management tools to position the data they collect as useful in application monitoring environments.

Given the attention in this sector generated by Cisco’s $3.7 billion acquisition of AppDynamics, 451 Research anticipates additional acquisitions and mergers between NPM and APM vendors.

Category Overlap

Code-level APM, RUM, synthetic monitoring, performance testing and NPM represent several emerging instances of a blurring of lines across capabilities, as well as combinations of product categories.

The most notable new confluence involves NPM and APM. While these tools have historically had discrete functionality and uses, NPM vendors have recognized the utility of their offerings in APM use cases.

While NPM tools offer insight into different layers of application infrastructure than traditional APM tools do, they can both deliver valuable data about performance that is useful to professionals tracking application performance.

Another emerging trend involves integrating performance testing products with a variety of other tools, including APM. This stems from changes in the approach to testing that shifts testing activity to earlier in the application development process, as well as embedding it throughout the development cycle.

The result is demand for integration that allows developers to leverage APM tools to examine the results of performance testing in order to help pinpoint the source of problems.

It is also increasingly common for individual vendors to offer APM, RUM and synthetics capabilities, since these tools are often used together to detect and troubleshoot performance problems.

Additionally, there is a growing connection between APM and infrastructure monitoring, with vendors combining the two services.

In some cases, when application monitoring is added to infrastructure monitoring tools, code-level insight isn’t offered, but the combination is still valuable in today’s increasingly software-driven world, where application insight is important to teams responsible for infrastructure performance.

Conclusion

The application and infrastructure performance market has developed into both a big-data problem and a new opportunity. Vendors’ embrace of software-as-a-service delivery models, machine learning and big-data back ends demonstrates a response to the growing volume of operations data. Opportunity is emerging as providers explore ways to introduce the business world to the value in the data collected by monitoring tools.

It’s a pivotal time in the application and infrastructure performance monitoring market, with vendors hustling to meet the changing needs of businesses.

Vendors that can adapt to these needs and supply the right tools for the job have the opportunity to increase the value of their tools to supply data that can enable smarter decision-making for both IT teams and entire enterprises


Rate this ➜

0 people like this.

Tags: , , , , , , , , , ,

No comments.

Leave a Reply

*