Infrastructure is becoming less tangible as virtualization and the cloud take over, but that does not mean it is inscrutable. Indeed, abstract architectures can still fall prey to bottlenecks and other detriments on the physical layer, but it requires a new breed of monitoring platform to pinpoint what is really going on and how to correct it.
According to Virtual Instruments, CIOs are well aware of the performance challenges that affect their newly virtualized and distributed architectures, but have yet to confront them in a meaningful way. In a survey of European enterprise executives last fall, the company found that a significant majority believes infrastructure performance management (IPM) can improve agility, reduce the risk of failure and lower costs, although only 13 percent have implemented the technology. And nearly a quarter of the rest say they have no plans to create one in the near future. A key challenge is the multiple layers of abstraction that accompany the typical virtual data environment, which can hamper visibility and interfere with proper resource provisioning.
To be sure, however, there is no shortage of infrastructure monitoring and management solutions entering the channel. A company called SignalFuse recently emerged from stealth with a SaaS-based platform that offers an analytics approach to application and infrastructure management. The SignalFx platform assumes much of the data gathering functions that are normally done manually, allowing IT staff to concentrate on interpreting results, identifying patterns and sharing insights. The system provides real-time multi-dimensional data that can be used to correlate performance with underlying infrastructure characteristics, while an historical analytics module delivers customized analytics from thousands of sources to chart percentiles, moving averages, growth rates and other aggregated projections.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=i
Oddly, says ServerWatch’s Paul Rubens, it can be more difficult to monitor smaller virtual setups than larger ones, given that most platforms are aimed at top-tier enterprises. However, virtualization consulting firms like Europe’s Xtravirt are offering specialized Reporting as a Service (RaaS) platforms that can be tailored to any size enterprise. The company’s SONAR offering provides baseline measurements, configuration reporting and other functions to enable companies to assume proper security and compliance postures for their industries and maintain stability in day-to-day operations. The platform utilizes a small clientless agent to query virtual infrastructure and deliver results to the SONAR engine on the Azure Cloud. Results delivered back to the enterprise consist of a 42-point infrastructure assessment, best-practice analyses, software and configuration issues and other data.
A key problem that many enterprises run into when applying old-line management tools to newly virtualized environments is the constant flood of alerts that are generated, most typically in environments consisting of multiple, disparate monitoring systems. To counter this, platforms such as the Galileo Performance Explorer have added new alerting functions and GUIs that filter the noise of multiple solutions. The Galileo system offers integrated health assessments across operating system and storage infrastructure and provides proactive troubleshooting for business critical functions. Issues that cannot be dealt with automatically are processed through a streamlined alerting mechanism so that warning lights flash only when true problems arise.
The basic conundrum of virtual infrastructure management is that it is very difficult to manage what you cannot see. Without extensive visibility into the multiple layers of abstraction that characterize most virtual deployments, the enterprise has no way of knowing where trouble is lurking until it is too late.
The rush to lift data operations off of physical hardware is understandable given the utilization and flexibility improvements that can be had, but failing to upgrade management and monitoring stacks to reflect this new environment is a recipe for disaster.
Virtual infrastructure is already quite extensive in most organizations, and delaying the modernization of the management stack serves only to keep a crucial element of enterprise performance rooted in the past.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata, Carpathia and NetMagic.