It’s no surprise that the COVID-19 pandemic supercharged the move to the cloud for organizations large and small. As a global Cloud Service Provider (CSP) delivering IaaS and PaaS solutions for over 20 years, we at OVHCloud saw this sea change as an opportunity to help our customers understand what they need to consider during the transition, since many had to do make this change at a very accelerated pace.
This shift is an ongoing process that is constantly evolving as customer needs and the technology do. To handle this monumental change, and have long-term success, CSPs need to offer a wide variety of solutions to support their customers through this journey. One way to approach this is to think of making a 3-layered cake.
The Base Layer – Resiliency
The base layer for success as every organization realized immediately is resiliency – essentially being available for your customers and being able to continue operations. As a result, companies have had to re-imagine and re-engineer their IT infrastructure to make sure all aspects of the organization – from application development, to CRM, SCM to delivery and billing – keep functioning without any pause in operations.
The result – many organizations have chosen to use a hybrid and multi-cloud approach to reduce risk by spreading infrastructure between on-premise and one or more clouds. The hybrid often consists of leveraging existing on-premise machines and using cloud-based bare metal servers, some public cloud and for enterprises private clouds.
The hybrid approach gives companies greater resiliency by spreading out workloads on to the best type of cloud platform. For example, companies often use a public cloud for continuous integration and deployment and testing, given the public cloud’s elastic and affordable nature. However, for heavy compute, AI and storage needs bare metal servers are ideal given their customizability and raw power. In cases where security, certifications and dedicated resources are essential, think handling financial or personally identifiable information, a private cloud might be necessary.
Infrastructure resiliency is now a strategic pillar for any organization. While there are still many organizations that have on-premise infrastructure (nearly 80% of legacy business apps are still on-premise because moving them to the cloud takes time, money and resources), having a solid infrastructure in the cloud, be it private or public, is now the baseline for organizational resiliency.
The Middle Layer – Cost Predictability
Having your infrastructure in the cloud is meant to give you a degree of cost transparency. Cost predictability, however, is a more difficult thing to achieve with some CSPs. There’s a lot to consider so you don’t end up paying for more than you need, and equally important, so you can clearly forecast what your costs will be. Unexpected bills are not what leadership wants to see.
Does your CSP charge extra for network traffic or is it included in the price? Do they charge for ingress and egress? What software and licenses are offered – out of the box – to help you monitor usage and run manage your infrastructure? Is Anti-DDoS part of the standard offering, ad if yes, is it at multiple layers (L2, L3 and L7)? Sometimes it’s difficult to get answers to the simplest questions.
The best way to achieve cost predictability is to look at the whole package when choosing a CSP – not just the pricing, but for every offering and option. Consider if there are additional licensing fees or are licenses already included. See what network traffic costs are (they should be included). Also determine what backup and DR options are offered and are they managed, and what PaaS options are available.
The Top Layer – Flexibility to Pivot
You now have the first two layers of your cake. But to make it all work, you need more than just a mix of on-premise, bare metal servers, or private clouds. You need the right software components to orchestrate all this, in order to gain the flexibility to pivot and profit when the next shift in the business landscape occurs.
Two essential components are employing a Hyperconverged Infrastructure (HCI) through your hypervisor layer to manage resources, policies and control costs, and app containerization solutions to handle the DevOps tasks of getting applications developed tested and deployed.
HCI has become the strategic base for dispersed workloads (compute, network, storage) across clouds. According to 451 Research[i] more strategic workloads, such as data analytics and application development, are increasingly moving into HCI platforms. Orchestrating all these loads is much easier with a solution like VMware’s vCenter and vRealize Cloud Management. It is the binder to automate, monitor and allocate resources. It is one perhaps the key ingredient to making everything work together.
Furthermore, as just about every organization or business now produces software to some degree, developing, testing and deploying applications continuously and seamlessly is an ongoing challenge. Use a managed Kubernetes platform so the flow from development to operations to deployment is as seamless and automated as possible.
Is There Icing on the Cake?
While managing your IT infrastructure in the cloud will always be complicated, it’s worth it. Not having to maintain your infrastructure is like having a bakery that serves up custom cakes. You get to choose your solution – and adapt it as needed – so your organization or company can focus on what it does best.
[i] 451 Research – Voice of the Enterprise: Hyperconverged Infrastructure Strategy And Workloads 2020