Industry executives and experts share their predictions for 2022.  Read them in this 14th annual VMblog.com series exclusive.

FinOps in the New Cloud Order – Efficiency Vs. Optimization

By Jeff Kukowski, CEO at CloudBolt

Cloud costs are devouring an ever-growing share of company budgets. According to a recent paper by Andreesen Horowitz, cloud inefficiencies are costing some companies up to 2x more than if they had simply stayed with their own data centers; the report estimates this to be a trillion-dollar issue in terms of valuation reduction.

In order to keep cloud spend under control, companies must understand the difference between cost efficiency and cost optimization. They sound similar, but they’re actually two different things.

Cost efficiency is a traditional approach to cost control. Companies look for places in the organization where money can be saved, and then act on it. This could be choosing one vendor over another, hiring contractors to perform certain tasks instead of expanding your in-house team, preemptively implementing security tools, training and procedures to prevent expensive breaches, and many other cost-saving initiatives.

Cost efficiency is important. But these types of measures are usually singular actions, and they often don’t fundamentally address the root causes of wasteful spending and cost overruns. Cost efficiency is reactive in nature, and doesn’t emphasize open communication and shared responsibility across teams.

Cost optimization, on the other hand, is an organizational culture of awareness, shared responsibility and automated processes that continually evolves to eliminate waste and overages. It is founded on automation, deep visibility and communication, and is far superior to manual, ad-hoc cost-saving measures.

The Limitations of Cost Efficiency

When human beings perform job functions, we aren’t usually thinking about the cost of execution to the company. Someone else has already budgeted for whatever cloud, tool or process we’re using, and we simply want to complete the task at hand. We are also imperfect; we forget and neglect things, especially when the consequences don’t directly affect us.

This leads to many scenarios that contribute to ballooning cloud spend:

• An engineer might spin up a development environment, use it for a week, then never turn it off.

• Virtual machines or storage in public clouds are spun up and then forgotten, turning into zombie VMs that drain resources.

• The attached storage from a spun-down VM can be left running unwittingly. This results in orphaned storage that runs but is never used.

• Public cloud resources are often left on when not in use, such as weeknights and weekends.

• New releases have to be scaled up to add more code, but developers might neglect to scale them back down afterward.

• Disaster recovery (DR) snapshots are critical, but they are expensive to maintain and get outdated fast. If they’re not deleted as soon as possible, money is needlessly wasted.

In a cost-efficiency approach, human beings with spreadsheets are responsible for manually checking and verifying these things. But people often procrastinate when they have more pressing tasks at hand, or just forget altogether. When commitments are neglected, wasted resources accumulate, and the result is wasted money. In fact, the top problem cited by FinOps pros in the State of FinOps 2021 report was getting engineers to take action on cloud spend – 39% said this was their number-one issue.

The Evolution to Cost Optimization

In an optimized, shared responsibility model, enterprises implement an advanced alert and visualization system that informs all stakeholders of the financial impact of the daily decisions they make. This requires 360-degree visibility across all clouds, the automation of many manual processes, and built-in permissions and guardrails. For example:

• Rather than multiple static views of siloed departments, holistic cloud visualization tools provide end-to-end visibility of all cloud resources and configurations across the organization.

• If an engineer forgets to dismantle a development environment when it’s no longer in use, an automatic notification system sends them a reminder to do so.

• Anyone provisioning public cloud resources is automatically notified of potential overspend, and less expensive options are recommended.

• Instead of relying on individuals to check recommitments and record results in a spreadsheet, the checking is done automatically every 24 hours.

• People don’t have to chase down others to get them to turn off unused cloud resources, because built-in guardrails turn them off after a certain amount of time.

A cost optimization strategy also proactively prevents security incidents and compliance issues. Just as organization-wide visibility and automated notification systems can continuously detect and remediate cost overruns, they can also secure your cloud infrastructure, both up- and downstream. For example, if a developer inadvertently writes a piece of code with security issues, they would automatically be notified of the vulnerability, and possibly given a safer alternative.

Furthermore, automating cloud resource deployment processes and configurations allow organizations to automatically check their cloud services against compliance standards such as PCI-DSS, CIS, and the AWS Well-Architected Framework. Currently, many organizations need to put in requests to compliance officers, then wait months for analyses to be completed. The entire time you’re waiting, you could be out of compliance or have security vulnerabilities and not know it, resulting in expensive breaches or fines.

“Good Enough” Is No Longer Enough

Having a thousand people asking others to take cost-saving measures and remember recommitments as opposed to policies and guardrails that execute automatically is simply not sustainable in the new cloud order. As hybrid cloud grows more expensive and complicated, these companies will be further mired in manual tasks, accumulated waste and cost overruns – not to mention sub-par compliance and security.

Until recently, a cost-efficiency mindset has been good enough, but “good enough” is not good enough anymore. In the new cloud order, the most nimble and competitive enterprises will be the ones that embrace a cost-optimization approach to their FinOps.

##

ABOUT THE AUTHOR

Jeff Kukowski is the CEO of CloudBolt. He has extensive experience scaling technology companies and category-leading solutions across all stages, including start-ups, VC-funded to exit, private equity-backed, and public company turnarounds. Previously, Jeff served as CEO and Board Director at SecureAuth, a leading cybersecurity and identity and access management (IAM) company. Prior to SecureAuth, he served as Chief Operating Officer, Chief Marketing Officer, and Executive Vice President of Sales at Axon (NASDAQ: AAXN), manufacturer of TASER. Earlier in his career, Kukowski held executive leadership roles at Destinator Technologies, Cyclone Commerce, and Cognizant. Jeff holds an MBA from the University of Chicago and earned his undergraduate degree at Northwestern University.