Once upon a time, the term cloud meant public cloud, and many commentators thought it would remove the need for enterprises to run their own servers or locate infrastructure in their own choice of datacenter. But over the past few years, the word hybrid has emerged not just as an aspiration, but as the model of choice for most enterprises.
Although public cloud offers great scalability and flexibility, the fact is that in this model, the service provider makes many decisions on the customer’s behalf; for enterprises that need to know how and where their data is being located, hosting on-premises is still a desirable option. Hybrid cloud gives enterprises the flexibility to choose where to place each workload depending on that individual workload’s requirements.
But how unified are these different venues today? Public cloud is by its very nature consumed in an on-demand manner, giving it a reputation for scalability and flexibility. On-premises infrastructure has traditionally been procured in a fixed capacity for a fixed price, which is inflexible compared with public cloud but does give reassurance of data security.
To better understand these facts, HPE partnered with 451 Research to examine the typical cloud experience. Our key findings include the following:
- Multi-cloud is a mainstream strategy being adopted by 57% of enterprises today, but public cloud and private are still different beasts, especially regarding scalability, cost optimization, and agility.
- Public cloud is a positive experience for 97% of the IT decision-makers surveyed globally in a range of industries.
- Despite the positive public cloud experience, on-prem deployments continue to be desired for reasons of security, data protection, and compliance; 68% of respondents said security was the biggest challenge in using public cloud.
- Many enterprises struggle with on-premises capacity planning and take a ‘better safe than sorry’ approach. Most are wasting 40-50% of server capacity, probably because 92% of enterprises have to wait a month or longer for new capacity to be installed. Considering that 83% of enterprises are still using fixed-capital models for on-premises capacity, wastage is difficult to prevent.
- For on-premises deployments, challenges related to scalability and agility are our respondents’ biggest worries: 80% said provisioning times were poor, 70% said developers complained of poor agility, and 70% said capacity planning was a headache. The traditionally fixed nature of on-prem deployments, plus the need to administrate and manage the infrastructure, impacts the ability to consume rapidly and to scale when needed.
- Although public cloud does go some way to aid capacity planning, there are still difficulties in buying the most cost-efficient instances; 71% claim administering public cloud reserved instances is a bigger challenge than expected.
- Decision-makers are between a rock and a hard place – they need to keep resources secure and compliant (for which on-prem deployments can provide a partial solution), but they also need to allow scalability, which is constrained by the fixed-capacity model, while reducing waste.
- Consumption-based pricing for on-prem deployments can provide the flexibility of public cloud with the reassurance of data control. In a hybrid cloud model, this gives enterprises a choice of venue for each workload, with the ability to scale on demand.
Click here to read the entire white paper.