The Fifth Annual State of the Network Global Study by Network Instruments, which does network and application performance monitoring, shows that the most popular method of cloud computing is software-as-a-service (SaaS), which is embraced by 67% of 163 IT professionals surveyed, followed by private cloud computing (49%), infrastructure-as-a-service (IaaS), also known as the public cloud (32%), and platform-as-a-service (PaaS) (15%).
Cloud computing is one of three areas the survey addressed, which Network Computing is reporting in a three-part series. Part one looked at the promise and problems of videoconferencing. The third installment will look at application management on computer networks.
The number one concern about cloud adoption is the security of corporate data, expressed as a challenge by 74% of respondents. Other concerns pale in comparison but are still significant. The lack of ability to monitor the end-user experience was raised by 37% of respondents, then compliance issues (33%), the impact on network bandwidth (32%), cost increases (29%) and a lack of interoperability standards (25%).
Further, when they were asked how cloud computing affected overall performance based on four criteria, most noticed little difference in cloud versus on-premise and, in some cases, performance declined. When asked about cloud service availability, most saw no difference and 33% saw an improvement while only 8% reported a decline. On the quality of the end-user experience, 25% noticed an improvement and only 12% reported a decline.
When asked about their ability to monitor their cloud service, only 20% saw an improvement while 22% reported a decline, and most detected no change. Lastly, when asked to rate their ability to troubleshoot problems with their cloud service, 28% reported it got worse, 55% reported no difference and 17% saw an improvement.
“When moving to the cloud, you have to relinquish some control,” says Charles Thompson, director of product strategy for Network Instruments. “You don’t own the infrastructure. You can’t touch the infrastructure. You can’t deploy analysis technology at a whim. And so, [there’s] that lack of a local tangible asset that you could interact with.”