Many times, it is characterised as a commodity and positioned as a rapid path toward innovation. IT expects cloud to give them fewer boxes to maintain, fewer hypervisors to operate, and even fewer purchasing decisions to make.
In reality, however, managing a cloud footprint is much harder and more specialised than many imagine. Cloud services must be selected, workloads must be migrated and usage must be tracked. Not unlike other complex IT systems, cloud infrastructures have to be monitored and managed, and clouds much touted easy scalability depends on specific infrastructure capabilities and a watchful eye to identify and correct problems.
Over the years, clouds have emerged that are built on proprietary platforms and optimised for the comfort and efficiency of the cloud providers. This has shifted the burden of learning how to operate and optimise these platforms onto the user. In fact, many cloud implementations require deep knowledge of scripting, an understanding of the strengths and weaknesses of the architecture, and at a minimum, a strong familiarity with the new platform.
Along with this, the pricing models of clouds have become a feat of financial engineering. This has compounded the burden on IT not just to operate the cloud footprint but to optimise its cost according to a set of remarkably challenging pricing models.
And many teams continued to be surprised by the challenges of cloud. In fact, in April 2014 when iland commissioned analyst firm EMA to conduct a global survey on real-world experiences with public cloud providers, 98 per cent of UK respondents reported they experienced at least one unexpected cloud challenge.
Among the survey pool of 400 professionals, the vast majority had made some foray into cloud, and what became apparent was that there were six primary areas where companies discovered unanticipated obstacles. Here are the top six unexpected challenges:
1. Pricing (38 per cent)
The top challenge unearthed in the survey indicated that while cost savings may be a key benefit of cloud, current pricing models under which they operate are difficult to understand. Organisations therefore need to carefully analyse pricing models and their own IT needs before they commit.
2. Performance (38 per cent)
Different clouds are architected with different back ends, and some are more susceptible to noisy neighbour syndrome than others. For organisations sensitive to variations in performance this can impact the cloud experience.
3. Support (36 per cent)
The realities of cloud support contracts often take companies by surprise. Simple email or ticketing support may only be available to organisations at lower tiers. Companies purchasing higher-end support may still have difficulties getting access to adequate levels of hands-on expertise. Poor or overly expensive support can become a grating factor limiting cloud success.
4. Downtime (35 per cent)
Though many perceive cloud to be immunised from outages, the reality is that failure does happen. Its important to understand service level agreements and business continuity options.
5. Management of cloud services (33 per cent)
Just like on-premise systems, cloud services require IT management. However, many cloud vendors focus on technology innovation instead of simplifying cloud service management. The cloud is a new platform to most IT shops and it can take some time to learn.
6. Scalability (33 per cent)
A top promise of cloud is scalability, as it is supposed to allow teams to scale up to meet demand and scale down when the peek has passed. However, most companies experience real challenges both in scaling individual workloads and scaling their entire footprint. Knowing when and how to scale is not always straightforward especially because some vendors require an environment to be shut down and moved in order to get the job done.
Dante Orsini is SVP of Business Development at iland.