This article is part of the Key Research Findings from the DZone Guide to Cloud: Serverless, Functions, and Multi-Cloud.

Introduction

For this year's DZone Guide to Cloud, we surveyed 739 software professionals from across the IT industry, asking them questions on various topics about cloud technology. In this article, we focus on the data we received around how developers actually use the cloud in their day-to-day work as programmers. 

VMs, CPU, Deployment, and Cloud Native

77% of respondents have deployed web applications in a cloud environment and 62% have deployed enterprise apps in the cloud. These proved to be the two most popular answers among respondents and will thus form the basis of our comparative inquiries in this section. But first, we need to understand how developers use the cloud.

When it comes to configuring cloud virtual machine instances, 69% of respondents optimize for CPU, 68% for memory, and 44% for storage volume. Given the steep drop between the percentages for memory and storage volume, it seems that respondents’ main concerns when it comes to cloud VM optimization are compute and memory. To house these functions, 61% of respondents told us their organization uses a single cloud, rather than a multi-cloud, environment. Once configured, a large majority of respondents use their cloud platforms fairly evenly between three stages of the SDLC: production/deployment (74%), development (65%), and testing/QA (62%).

If we compare these numbers to those for respondents who have deployed web and enterprise applications in the cloud, a trend of unanimity appears. For respondents who have deployed web applications in the cloud, when it comes to configuring cloud VMs, CPU (75%) and memory (74%) proved by far the most important. And for respondents who have deployed enterprise applications in the cloud, again, CPU (77%) and memory (75%) took the cake. Thus, it appears that across application types (for our respondents, at least), the cloud’s most powerful and pertinent characteristics are the computing power of the virtual processors contained in a particular cloud solution as well as its ability to store large amounts of data.

When it comes to developing applications specifically for the cloud, we see much the same pattern. When asked if their company develops cloud-native applications, 46% said “yes” and 41% responded “no.” For those developers who have deployed web applications on the cloud, 50% work for organizations that develop cloud-native applications; and 52% of respondents who have deployed enterprise applications on the cloud work for organizations that create cloud-native applications. Interestingly, despite these fairly high numbers in favor of cloud-native development, the percent of survey-takers who use cloud development to target mobile device clients proved low. 39% of respondents told us that between 1-25% of their cloud development efforts are geared toward mobile, and 21% said none of their cloud dev efforts are aimed at mobile clients.

Conclusion

Now that we've seen how developers prefer to use the cloud, in our next post we'll discuss the platforms, frameworks, and container environments our respondents love. 

What's your favorite part of developing in the cloud and what aspect of developing in the cloud just drive you crazy? We'd love to hear down in the comments. 

This article is part of the Key Research Findings from the DZone Guide to Cloud: Serverless, Functions, and Multi-Cloud.