For enterprises, managing cloud costs has become a huge problem.
In the RightScale 2018 State of the Cloud Report, 80% of enterprises surveyed cited managing cloud spend as one of their key challenges. And, the more experience organizations had with cloud computing, the more likely they were to point to costs as an obstacle.
Part of the reason why cost management is so challenging is because organizations are spending a lot of money on public cloud services. More than half of enterprises (52%) told RightScale that they spend more than $1.2 million per year on clouds services, and more than a quarter (26%) spend over $6 million. That spending will likely be much higher next year, as 71% of enterprises plan to increase cloud spending by at least 20%, while 20% expect to double their current cloud expenditures.
Given those numbers, it's unsurprising that Gartner is forecasting that worldwide public cloud spending will "grow 21.4% in 2018 to total $186.4 billion, up from $153.5 billion in 2017."
Another problem that contributes to cloud cost management challenges is the difficulty organizations have tracking and forecasting usage. The survey conducted for the SoftwareONE Managing and Understanding On-Premises and Cloud Spend report found that unpredictable budget costs was one of the biggest cloud management pain points for 37% of respondents, while 30% had difficulty with lack of transparency and visibility.
Similarly, 40% of the IT decisionmakers surveyed for the 451 Research Voice of the Enterprise: Cloud Transformation, Organizational Dynamics report said that cost savings was one of their primary motivations for moving to the cloud. However, after their cloud migration, 53% said that cost and budget were still a key pain point for them.
In order to eliminate these pain points, enterprises are looking for ways to better manage cloud costs. In the RightScale survey, 58% of respondents said they planned to optimize cloud costs in 2018, making it the number one cloud initiative for the year. A Forrester Report found that 45% of enterprise IT decisionmakers surveyed have taken steps to reduce cloud spending, and another 37% have projects in the works.
So what steps should these organizations be taking? The following slides detail 10 tips that experts, such as analysts from organizations such as Forrester, Gartner and 451 Research, offer for managing cloud costs.
1. Monitor spending on a daily basis.
The ease and speed of provisioning new computing resources is one of the key reasons why organizations choose to use public cloud services. In most enterprises, developers and other IT professionals are spinning up new instances several times a day. If cloud managers are monitoring usage on a weekly or quarterly basis, they can easily get blindsided by unexpectedly high charges. That's why most experts recommend using a dashboard that allows you to monitor cloud usage on at least a daily basis, and preferably more often.
Some organizations choose to invest in third-party tools for this purpose, but the cloud vendors themselves also offer monitoring tools that enterprises can use for little to no cost. For example, Amazon Web Service (AWS) has a Cost and Usage Report, which tracks use on an hourly or daily basis, and a Trusted Advisor tool that helps businesses optimize their costs, as well as improving performance and security. Microsoft Azure offers Azure Monitor and Cost Management services, and Google Cloud Platform offers Stackdriver Monitoring.
2. Shut down unused or unnecessary instances.
Parents often remind their children to turn the lights off when they leave a room, and those reminders seem doubly necessary when it comes to public cloud computing resources. IT workers often spin up a new instance for a short-term purpose and then forget to turn it off when they are finished.
The RightScale survey found that enterprises are turning to both automated tools and manual processes to combat this problem. For example, 72% of those surveyed had taken steps to eliminate inactive storage, 71% of organizations shut down workloads after hours, and 69% specify expiration dates for workloads.
3. Require tags company-wide.
While some vendors have rolled out cloud management solutions that don't require the use of tags, most enterprises are still relying on tools that require them to tag their workloads. These tags can indicate a number of different attributes, for example, whether it is a Dev or production workload, which application or service it is for, which compliance requirements it must meet, when the workload should be shut down, and which department should be charged.
But these tags are only helpful if everyone uses the same tags and uses them consistently. Experts recommend that organizations clearly communicate the need for accurate tags to all cloud users and run reports that show any workloads that haven't been tagged. Organizations might also want to consider automation tools that require workloads to be tagged and issue immediate alerts or shut down untagged workloads.
4. Rely on automation rather than manual processes.
As their use of the public cloud grows and they begin using more cloud providers, many organizations find that the manual processes they used to monitor and optimize public cloud usage in the early days are no longer adequate. Automated solutions make it easier to track usage and enforce policies company-wide. Forrester analyst Lauren E. Nelson has noted, "Many organizations turn to the cloud to reduce spend on infrastructure or avoid steep upfront costs for new investments. Still, many have not achieved these goals. Success depends more on the maturity of cloud management and governance practices than the nature of the workload. With cost complexity continuing to increase alongside growing usage, users, accounts, and instance types, [infrastructure and operations] professionals increasingly depend on tools to enable visibility, consistency, and scalability of management practices."
5. Consider a standalone cloud cost management tool.
For many enterprises, the type of automation solution that might make the most sense is a standalone cloud cost monitoring and optimization (CCMO) tool. These solutions often cost just a fraction of the price of more complex hybrid cloud management tools. And according to Forrester's Nelson, "CCMO customers quote ROI at between two and six months, which is far quicker than more costly tools that also cover compliance, orchestration, and other solutions."
Quite a few different vendors have rolled out CCMO tools. In its Q2 2018 Wave report, Forrester listed RightScale, CloudHealth Technologies, Turbonomic and Densify as leaders; Apptio, CloudCheckr and Microsoft as Strong Performers; and Cloudability and Teevity as Contenders.
6. Invest in a hybrid cloud management tool if your needs are more complex.
Larger enterprises with more complex needs might find that they need a hybrid cloud management (HCM) solution. These tools often combine monitoring and cost management capabilities with compliance, orchestration, brokerage, template creation and additional automation capabilities. HCM solutions are more expensive but also offer a much broader range of capabilities than CCMO solutions.
As in the CCMO market, many different vendors offer HCM solutions. In its most recent Forrester Wave report on HCM Tools, Forrester listed RightScale, Scalr, VMware and Embotics as Leaders; Cisco, CloudHealth Technologies, Turbonomic, HyperGrid, Micro Focus, Red Hat and BMC Software as Strong Performers; and RackWare as a Contender.
7. Look for a solution with machine learning capabilities.
If you choose to use a CCMO or HCM solution, you might want to consider a tool that incorporates analytics and machine learning. This technology helps the management software get better at making cost optimization recommendations over time, allowing organizations to reduce their costs as much as possible while reducing the burden on IT staff. Nelson wrote, "Eventually, many CCMO vendors want their customers to trust recommendations by default, choosing to automatically implement recommendations in real time to more frequently squeeze incremental savings out of workloads. Such developments will fundamentally change the role of the cloud manager and these tools."
8. Take a systematic approach to cloud cost management.
Of course, cloud cost management requires far more than just technology — it also requires that organizations get the right people and processes in place. Gartner recommends that organizations approach the issue systematically. It offers a five-step framework for public cloud cost management:
9. Watch out for vendor lock-in.
Experts also recommend that enterprise cloud managers be very cognizant of the issue of vendor lock-in. The world's largest providers dominate the public cloud market. According to the latest data from Synergy Research Group, Amazon Web Services (AWS) dominates with a 40% share of the worldwide public cloud market, and the top three — AWS, Microsoft and Google — together control more than 60% of the market.
In this environment, enterprises should be wary of becoming overly dependent on any one provider or of using services that might make it difficult to migrate to a different vendor. These situations could drive up costs over time.
10. Optimize private cloud costs.
Finally, it's worth noting that enterprises aren't having trouble managing only their public cloud spending — private cloud costs remain challenging as well. The SoftwareONE report found that 56% of respondents felt that public cloud and in-house data centers were equally complex and time-consuming to manage, and 38% said that public cloud and on-premise pricing were equally complex and confusing. According to a 451 Research white paper, 35% of enterprises are moving data back from the public cloud to the private cloud. But moving from the public cloud to private cloud may not resolve cost management issues. 451 Research wrote, "Our recommendation is to use tools, outsourced services and partnerships to optimize private cloud as much as possible to save money, while still benefiting from the more value-adding benefits of private cloud, such as control, performance and security."Cynthia Harvey is a freelance writer and editor based in the Detroit area. She has been covering the technology industry for more than fifteen years. View Full Bio