The complexity of choosing and managing cloud long-term leads to tough decisions for most modern enterprises. One industry expert notes that choosing cloud doesn’t have to be the default for enterprise IT teams. Some applications just aren’t cloud-compatible and may never be. An on-premises virtual environment might be a better fit for a mission-critical application. Cloud platforms depend on the idea of applications sharing resources, and not every enterprise application was built that way. The decisions around what type of cloud will be appropriate for each application will depend a lot on the size and importance of each app. It’s also going to depend heavily on what might be needed in six months or a year from each application. Enterprise IT teams should consider microservices as the future of application delivery.
Cloud costs also play a huge role in any decision. Recent research from 451 Research delves into the realities of cloud costs, and what enterprise IT teams can do once they’re armed with this knowledge. There’s a typical public cloud expenditure curve that happens that ideally will end in cloud adding value without adding excessive costs. It’s a useful way to think about the costs of cloud, and when in the lifecycle of a cloud application those costs will affect your organizational and IT budgets. For example, the cost of a new application in the cloud may be very low, as developers or users play around with it. Then the cost increases as it grows without oversight. In contrast, switching an application in use to a cloud-based model will require up-front costs (whether IT’s time or subscription or licensing fees) but may prove to be cheaper over time. This research found that after a cloud migration, costs are the top pain point, with 53% of respondents naming that. Security concerns comes in second. There’s also interesting data in here about the number of companies that bring workloads back in-house after trying them in the cloud, as well as the various levels of policies and governance in place for cloud apps. Successful cloud adoption really does involve playing the long game.
On the topic of managing cloud deployments, here’s a look at how cloud has totally transformed the role of compliance at any organization. It used to be a lot easier to know which workloads were running where when everything was on-premises. Along with the servers and applications themselves going through a huge shift, so have job roles and policy requirements at companies. Part of the challenge to maintaining continuous compliance in the cloud is the quick-changing nature of cloud, along with the risk introduced by using off-site vendors and sites out of the control of IT. In addition, cloud locations vary widely, and multicloud environments are becoming common. We happen to be able to recommend an excellent monitoring solution to help you regain the visibility lost with cloud adoption and get ahead of all the associated issues.
At the heart of all these cloud deployments is a whole bunch of hardworking servers. These servers got a boost last week with announcements from HPE and IBM, which both released AI-oriented servers. HPE said its new server product performs three times better than the previous version, and is designed for high-performance computing as well as deep machine learning use cases. And IBM said in its announcement that more than 300 members of its processor consortium are developing AI-based products and services. The future is coming at us fast.