Use these evaluation criteria to help you determine whether and how the cloud might benefit your organization – and at what operational cost. We’ll assess both Public and Private Cloud offerings, and explore how each addresses the requirements and constraints of different organizations.
This blog was originally published here.
To cloud or not to cloud: that is the question. A critical component of a good cloud strategy is a defined process to determine if a service should be brokered or consumed directly from a public cloud provider, or built out internally as part of a private/hybrid cloud environment. Enterprises need to define the evaluation criteria to enable the intelligent use of public cloud services while maintaining appropriate governance and control.
In “The Importance of a Cloud Strategy” I discussed why a good cloud strategy is a requirement for a successful IT transformation initiative. I shared how many organizations will state that they have a “cloud-first” strategy, but without a common understanding across the organization of what that means or how it will be achieved.
Some organizations have defined “cloud” as meaning “public cloud” services only. Others use “cloud” to refer to the adoption of the “cloud operating model”, with services hosted on both public and private/hybrid clouds. Others associate “cloud” with “cloud-native”; applications built to take advantage of cloud technologies. The cloud strategy must define what “cloud” means in the context of your organization.
I recommend aligning the cloud definition with the adoption of the “cloud operating model”, since the benefits to the organization come from the capabilities provided by the ecosystem, and not necessarily from where the services or infrastructure are hosted. The adoption of a cloud operating model to support private/hybrid and public cloud environments offers the most benefits and flexibility to an organization that needs to rapidly deploy new services to support the growing demands on IT from the lines of business.
The Cloud Operating Model
In an upcoming post “The Cloud Operating Model” I show how the adoption of best practices from the hyper-scale cloud providers can mature your IT organization, transforming from a reactive, resource-constrained cost-center to a proactive, responsive provider of IT services to your developers and partner lines of business.
This model changes the way IT services are built, governed, supported, consumed and managed. It allows organizations to provide consistency across IT services – from user experience to governance and management.
IT Service Delivery
In “Mature the Delivery of IT Services” I shared how to take existing IT services and “cloudify” them – with full automation for service provisioning, self-service access for consumption, and lifecycle management of the service.
This process should include an evaluation of whether the IT organization should build and host the service as part of a private/hybrid cloud environment, or broker access to the capability provided by a public cloud provider. In many cases the consumer does not care where the service is hosted or how the underlying infrastructure is configured – the consumer is focused on the capabilities that will allow them to be more productive in their own roles – developer or business user. It is the other cloud operating model stakeholders (IT, security, compliance, finance) that need to determine what makes the most sense for an organization when determining how the services will be delivered.
The cloud strategy should provide clear guidance on the criteria used to make this determination, which will be dependent on the requirements and constraints of the larger organization.
Public Cloud Drivers
The promise of public cloud – agility, flexibility, cost-savings – has delivered in many areas but not all. Like everything in life, for every benefit there is likely a corresponding drawback. How suitable public cloud is for an organization is dependent on the priorities, requirements, and constraints of that organization.
Private Cloud Drivers
Private cloud is often seen as “legacy”. This is in part because many organizations labeled their virtualized environments as “private cloud”, despite not having many of the capabilities typically associated with the cloud operating model, such as automation and self-service.
Virtualization != Cloud
Virtualization is a foundational requirement for cloud, but additional capabilities, such as those provided by the vRealize Suite, are required to deliver a true private/hybrid cloud aligned to the cloud operating model.
Enterprise organizations are quickly discovering that there are many roadblocks on the journey to full public cloud utilization; not least security, regulatory compliance and audit, but also often a lack of understanding for existing applications to be migrated or refactored. Moving IaaS or VMs to a public cloud provider (“lift-and-shift”) rarely adds significant value and more often reduces visibility and increases risk and cost. Few organizations are able to successfully complete a wholesale migration of workloads to a public cloud and so it makes sense to design a private/hybrid environment to provide the capabilities demanded by the developers and lines of business within the relative safety of your existing datacenter/resources/processes/teams – at least until the maturity of the organizations public cloud support model meets the minimum set of supporting functions required for performance, scalability, high availability, and business continuity.
The cloud strategy should include an overview of the decision criteria for cloud usage and should direct readers where to go to find the detailed, maintained criteria. The Cloud Architect or Cloud Center of Excellence should be responsible for collaboratively creating and maintaining the criteria and decision tree.
- Service placement – The decision criteria should provide guidance on where to host or create a new service. Services offer full lifecycle management of the provisioned resource and are consumed via a global portal or API. For example, Database-as-a-Service, automated Kubernetes cluster provisioning, or Infrastructure as Code (IaC) full-stack application provisioning. For maximum control and visibility:
- Workload placement – The decision criteria should determine the placement of resources deployed by the services. Resources can be VMs, containers, or entire applications including security components. For example, development resources can deploy to AWS or Azure, PCI regulated applications and data can only deploy to the PCI-compliant environment within the private cloud. Note: Once the decision criteria for workload placement have been defined the vRealize Suite can fully automate the initial placement of the resources and ensure that the placement constraints are not violated over the life of the workload.
Multi-cloud (the mixture of multiple public and private/hybrid services) provides the most flexibility for your environment. To maintain effective control over these mixed environments organizations can utilize the vRealize Suite for operational consistency – consistent operations, consistent governance and policy, and a consistent user experience. VMware Cloud Foundation brings consistent infrastructure and operations across private and public clouds for a true hybrid solution, allowing maximum portability and flexibility. Together they allow organizations to provide a platform to support all of their workload needs with the flexibility to host services and resources where they can achieve maximum benefit with minimum risk.
Review the information linked below for further guidance on using the cloud to accelerate your organizations IT transformation initiatives:
- Keep your cloud strategy simple, flexible and effective with this easy to use guide – The Importance of a Cloud Strategy
- Automate like a Cloud Service Provider – Take the next step in the journey from IT Operations to IT Service Provider with automation and orchestration –
- Part I – Evolve Your Automation Journey
- Part II – Mature the Delivery of IT Services