Organizations are continuing to waste money on cloud resources that they do not require, with 50% admitting that overprovisioning of cloud resources was a major factor.
Idle or underused resources came just behind and a lack of skills was blamed by 43% of respondents to a survey.
Other factors included “temporary” cloud resources that are never ended, such as a service trialed by a customer that is never switched off, along with varying geographical costs resulting in more spending than expected.
The report from HashiCorp and Forrester, released this week, indicated the issue of overspending has not changed from last year’s figures.
Overprovisioning occurs when a customer reserves resources in the cloud, such as CPU or storage, but doesn’t actually use them.
Solving the cloud-native app puzzle with CNAPP
The value of integrating cloud-native application protection into security and development
While overprovisioning may simply appear like bad management or a loss of administrative control at first glance, the increasingly complex nature of cloud tooling can lead to customers buying services that are not necessarily needed.
While it might be possible to dynamically allocate resources in some instances - serverless architecture is a good example - this can result in performance or capacity issues, meaning reserving resources in advance is often unavoidable.
One solution to the challenge of overprovisioning is careful monitoring of actual usage in order to optimize cloud spend and gain visibility over one’s estate.
The adoption of best practices across an entire organization, such as using automation and infrastructure-as-code, can also cut down on overprovisioning through the implementation of well-defined processes around resource usage while also improving scalability.
HashiCorp’s report noted that the approach of adopting best practices also lends itself well to a multi-cloud environment. Almost all of the organizations surveyed that had done so said that multi-cloud was working well for them, or would be shortly.
The potential difficulties in adopting those practices should not be underestimated, though, the report read.
An issue that pervades the IT industry, 27% of respondents said skill shortages was one factor complicating the operationalization of multi-cloud in their organization. Siloed teams and a lack of training were also concerns.
Once best practices are in place, or well on the way, organizations are then well-placed to take advantage of savings by selecting services from a variety of vendors and avoid being locked-in to one particular cloud.
While avoidable cloud spend appears static compared to the previous year, the report concluded organizations that are further along the journey to the adoption of best cloud practices, including the creation of a platform team, reported a reduction in overprovisioned, idle, or underused resources as contributing factors.
