How to Avoid Runaway Costs in the Public Cloud

Lynn LeBlanc is CEO of HotLink.

The speed at which the enterprise has transitioned from a “cloud-never” to a “cloud-first” posture is astounding. The fear, uncertainty and doubt that gripped the industry when the public cloud was new and exotic has been replaced by a mindset that is already thinking past the mere deployment of cloud resources to the steps needed to optimize them. But the cloud is not the local data center, and both the tools and the techniques needed to keep a lid on costs are dramatically different when you rent infrastructure instead of owning it outright.

According to Gartner, the consumption of public cloud services is set to grow 18 percent in 2017 to top $246 billion. The research firm says this will represent the height of growth for the remainder of the decade, although the ensuing years will feature only a mild tapering. Among the leading software vendors, however, we can expect to see cloud-only releases approaching 30 percent of all products by 2019. Perhaps most interesting is that with the rationale behind the cloud quickly shifting from simply off-loading traditional IT workloads to supporting fully cloud-native applications and services, the need to optimize consumption will take center stage.

But how, exactly, can you manage something that you don’t own? For one thing, the cloud’s pay-as-you-go pricing model and the ability for line-of-business managers to circumvent IT to craft their own data environments makes it challenging for IT to keep up with what’s happening in the cloud, let alone control the costs. And even if planned cloud deployments do come in well under budget compared to standard data center infrastructure, there is no guarantee that costs will not spiral out of control once the workloads scale three-quarters particularly if usage monitoring and management are an afterthought.

This is the primary reason why monitoring and optimization need to be embedded as core principals in any enterprise cloud strategy. Only through continual, active engagement in the cloud’s infrastructure, abstraction, service and data layers can organizations gain sufficient leverage over resources, workflows, connectivity and all the other facets of cloud operations that define both costs and performance. This means the enterprise needs to determine how instances are being deployed, how applications are used, what will constitute effective monitoring, whether workloads are conforming to budgetary requirements and a host of other concerns. As well, advanced analytics tools will need to track multiple data points, such as instance sizes, usage patterns, update requirements, network bandwidth utilization, data retention policies and both historical and predictive analyses.

Since most corporate IT shops continue to have a dominant on-premise footprint, the integration of internal and external resources under hybrid management architectures is the logical step to manage the resources spanning public and private IT infrastructures. Many organizations, in fact, are working toward linking their local VMware environments to those provided by hyperscale providers like Amazon Web Services (AWS). By incorporating the management of public resources into the same software that governs the existing data center, IT gets a holistic management stack without adding operational or integration complexity. Not only does this provide full cloud control using the same management interface and workflows that IT has grown accustomed to, it provides a host of benefits like access to AWS’ GovCloud for applications that deal with public sector services, unified network and security management under VMware vCenter, and hybrid automation and orchestration under VMware’s PowerCLI stack. The enterprise also gains the ability to automate bi-directional workload conversions and migration for bursting to the cloud and other temporary use cases.

Most cloud experts are quick to say the true benefit of the cloud is not its low cost but its increased flexibility and the support it lends to digital transformation efforts. This is true enough, but costs are not an insignificant factor either, particularly as the cloud sets new lower benchmarks as to what the enterprise should be spending on data infrastructure.

Increasingly sophisticated management techniques allow the enterprise to govern its entire data footprint on a highly granular level, and in turn this helps the enterprise determine the most effective, productive means to leverage its cloud-based resources, as well as the prevention of runaway costs in the public cloud.

Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Penton.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.