CA Technologies has generated research on the new challenges facing performance monitoring. These challenges stem from complex computing, such as cloud computing and hybrid computing.
The frustrating effects of network outages and application malfunctions highlight the need for performance management tools to evolve for cloud computing. Performance management describes the tools and processes in place to detect and resolve any performance hiccups that might prevent end-users from using the network and/or application. There are three types of performance management tools: application performance management (APM), network performance management (NPM), and unified performance management (UPM).
CA Technologies’ performance management research focuses on pain points and methods to improve performance. Here are some of its key findings and advice on choosing a performance management solution.
CA Technologies Performance Management Tips and Advice
Already complex networking environments will become increasingly more complicated in the near future, especially when considering the advancement of the Internet of Things (IoT). According to Gartner, “By 2020, IoT technology will be in 95 percent of electronics for new product designs.” It’s common for network administrators to employ several APM and NPM tools to evaluate traffic flow. With the anticipated amount of IoT devices, CA Technologies suggests that it’s best to use a converged approach instead of using multiple performance management tools. In a joint presentation with Viptela, now part of Cisco, CA Technologies outlines the benefits of UPM. The upsides to UPM are that it provides a single pane of glass, centralized analytics, and shared core services and workflow across all elements in network and application monitoring.
CA Technologies’ Peter Waterhouse describes the current computing era as the “golden age of applications.” He points to the dynamic assets applications bring daily to human existence, such as healthcare and productivity. He also tells app developers and IT administrators that “75 percent of consumers expect a consistent experience whenever they engage” in applications. The present methods to keep up with these high expectations amount to implementing APIs, microservices, containers, and orchestration. These additions to the computing experience increase the difficulty for performance management tools targeting the problems affecting the network and applications. The tools take on a more reactive stance instead of a proactive one when it comes to monitoring dynamic computing environments.
To help network administrators decide which performance management solution to employ, CA Technologies coined the acronym FITPAL. The acronym stands for Fault, Inventory, Topology, Performance Application, and Logs. The company states that these components are essential to full visibility, and it lists the reasons why as:
“Fault: Capturing fault data becomes more valuable when it can be correlated to performance issues in a ‘single pane of glass.’
Inventory: In hybrid or multi-cloud environments, it is imperative to have a good handle on physical, virtual, and logical network elements. With infrastructure inventory data coming from multiple data sources, it is easy to lose track of what’s where. Ultimately, you can’t manage what you don’t know.
Topology: Using topology mapping, network managers can determine if neighboring infrastructure is causing a performance impact on a given application, providing insight into if or how routing should be adjusted, surfacing a hidden problem.
Performance: There are many aspects to performance data in heterogeneous environments. Modern multi-protocol networks demand that NetOps track SNMP, API, packet, and flow data, often from multiple vendors — in real time.
Application: Since the user experience is the most important metric, having the ability to triage application experience issues back to network infrastructure is a critical requirement for networks today and in the future.
Logs: Having the ability to perform log analytics against the other five FITPAL elements provides predictive capabilities for fault, performance, and capacity planning.”