Why Performance Monitoring Is Harder in the Cloud

    July 19, 2016

    By Luis Colon

    Transitioning to the cloud involves more than migrating systems — companies have to reevaluate their means for measuring success.

    It’s easy to track the growth of the public cloud market — $204 billion this year, up 16.5% from 2015, says Gartner — but it’s more difficult to track how companies grow within the cloud. As with applications and services, your performance monitoring tools can’t just be cut and pasted into a cloud environment. After they’ve migrated, many organizations are finding it difficult to maintain visibility into their operations.

    In general, charting performance is never completely straightforward; it requires IT professionals to gather many data points from a complex array of interrelated apps, servers, databases, and the like, often with competing operating languages and standards — a group of systems, or “IT resource,” rather than a single app. This can be difficult enough on your on-premise infrastructure, but when you move your systems to someone else’s virtual environment, the challenge becomes all the greater.

    Hurdles to Effective Monitoring

    As TechTarget observes, companies often find that the performance monitoring capabilities offered by cloud services vendors are not as robust as what legacy systems might have offered them.

    One reason for that is that vendors are now the gatekeepers to much of that performance information, and it’s up to their discretion whether to give or withhold it. On the other hand, many cloud services providers are relatively new, and many haven’t had sufficient time to build out performance management systems. This is one reason why you should always select a vendor with performance in mind.

    But solid information or not, purchasing a PaaS or IaaS service means that an organization will now have to monitor data both from their existing infrastructure and from their cloud provider. Unfortunately, those respective management systems aren’t likely to be entirely compatible. While organizations bridge the gap, usually with platform-agnostic interfaces or creative orchestration, IT professionals are faced with the task of unifying disparate systems under a single performance metric. As you can imagine, that work is far from easy.

    Leveraging Performance Tools

    In response to these dilemmas, many organizations are seeking assistance from third-party performance management vendors. TechTarget notes that while performance monitoring issues once plagued traditional IT, diagnostic tools helped to fill the void — once for private virtualization, and again today for private cloud. While these have helped to reduce the monitoring burden, at least in the cloud, IT still has to report on the entirety of the enterprise, and many software options are too niche to cover such a wide scope.

    A major impetus behind developing TeamQuest’s Vityl suite of performance monitoring and capacity management products was to grant organizations a comprehensive understanding of their enterprise IT environment, encompassing cloud, on-premise systems, or both. If organizations use tools with limited application, e.g. for one platform but not another, that does little to reduce their workload.

    The strength of our tools is that accurate figures of heterogenous performance are displayed in a single view, as simple measures of health and risk in our TeamQuest Dashboard.

    Although many organizations have sought to reduce their in-house IT spend by migrating to the cloud, it plainly introduces an increased need for performance management, at greater difficulty. At this point, the best thing enterprises can do is employ sophisticated monitoring tools to ensure that they are indeed getting the most cost-efficient performance out of their investments.

    (Image credit: Nicolai Traasdahl Tarp/Unspla3sh)

    Category: cloud-computing