Cloud Costs Too High? Poor Design Is Likely the Cause

Managing cloud spend remains a top challenge for organizations of all sizes, according to the Flexera 2023 State of the Cloud Report. Cost concerns were cited by 82 percent of survey respondents, surpassing security for the first time in a decade to become the No. 1 challenge.

Organizations exceeded their cloud budgets by an average of 18 percent, up from 13 percent the previous year. They estimated that 28 percent of their cloud spend is wasted, although that number is likely low.

Many factors play a role in cloud budget overruns. If an organization lacks effective policies governing cloud provisioning, developers may specify more resources than the application being deployed needs. Development and test environments may run 24x7 even though developers only work during business hours. Resources may be scaled to support maximum workload requirements instead of using auto-scaling features to handle spikes in demand.

Interested In Learning More?

Bad Design Equals Higher Costs

In many cases, however, poor design is the culprit. In fact, bad design can more than triple cloud costs. Organizations tend to treat the cloud the same as their on-premises environment, and fail to take full advantage of the available resources and tools.

Few organizations are aware that how you deploy your application can have a significant impact on the cost of that application. For instance, if you have an application that connects back to a data source in a different region, the bandwidth cost of the connection can be astounding.

The cost will be significantly less if you have a local cache copy of the database. The cache will also improve the application’s performance. Instead of accessing all of the data within the organization, you could limit the cache to the subset of data that’s relevant to the application.

Getting the Right Reporting

The Flexera study also found that less than half of organizations are using automated utilization monitoring to optimize cloud costs. Cloud providers provide dashboards that recommend ways to reduce your spend. However, these dashboards are only skin deep because there are features that cloud providers don’t turn on.

The biggest gap is that the server provider dashboards can’t tell you if a system is overprovisioned unless you turn on application-level monitoring and reporting services. You can’t tell if a particular VM is only using 10 percent of its CPU cycles or half of its memory on a regular basis. DeSeMa can turn on and tune those features so that you get deep analytics into the application.

We can also set up reporting on the application’s internals so that you know where you can control your cloud spend. Our reports focus on the last 30, 60, 90 and 365 days of application usage. We will tell you how the application is being used so you can decide how to scale the available resources. You might also find that you need to set up dynamic scaling to bump up resources at certain times of the year.

Reducing Security Costs

Security is another area where organizations overspend on cloud resources. Few organizations understand how to set up zero trust networking. Many ancillary security controls are designed to patch holes in an open network. In a true zero trust environment, you don’t need those controls because we make it impossible for certain types of communication to happen.

For instance, there’s no need to do port scanning in a true zero trust environment because the servers sitting next to your VM can’t talk to any of your ports. You can reduce costs by eliminating that tool and the administrative overhead associated with it.

In our next post, we’ll discuss how the right controls and automation can dramatically reduce costs by reducing cloud waste. Meanwhile, contact the cloud experts at DeSeMa to discuss your particular environment and cost concerns.