The problem with defining the difference between hybrid cloud and multi-cloud is that these two terms are often used interchangeably. However, for the similarities, there is one major difference.
In a multi-cloud environment, an enterprise utilizes multiple public cloud services, most often from different cloud providers. For example, an organization might host its web front-end application on AWS and host its Exchange servers on Microsoft Azure. Since all cloud providers are not created equal, organizations adopt a multi-cloud strategy to deliver best of breed IT services, to prevent lock-in to a single cloud provider, or to take advantages of cloud arbitrage and choose providers for specific services based on which provider is offering the lowest price at that time.
Hybrid cloud computing differs from multi-cloud computing in one significant way: the inclusion of private cloud infrastructure such as an enterprise’s own data center along with one or more public cloud services, usually working in conjunction to achieve business goals.
Thus the two major differences are as follows:
- Hybrid clouds always include a private cloud and are typically managed as one entity
- Multi-clouds always include more than one public cloud service, which often perform different functions. Multi-clouds do not have to include a private cloud component, but they can, in which case they can be both multi-cloud and hybrid cloud.
Many organizations adopt a multi-cloud strategy by accident, for example when different departments throughout the organization utilize different public cloud providers for a given function, while others develop a strategy for utilizing multiple public cloud providers as part of an all-encompassing IT strategy that includes on-premises, public-cloud based infrastructure as a service (IaaS), and SaaS offerings as a comprehensively managed hybrid IT environment.
For example, a hybrid cloud application might utilize on-premises databases while running application code both in the on-premises private cloud and ‘cloud bursting to the public cloud when demand increases beyond the capabilities of in-house hardware.
As technologies change, enterprises will continue to adopt multiple clouds to achieve their business goals, whether or not they are ‘born in the cloud’ or migrate to cloud providers over time. According to Gartner, anything other than a cloud-only strategy for new IT initiatives will require justification for over 30% of large-enterprise organizations, and they further predict that over half of enterprises that use cloud will develop a cloud-first approach by 2021.
Each public cloud provider has their own strengths and weaknesses, and each has their own cost structure. Developing a cloud strategy involves much introspection – assessment of existing workloads, databases, networks, SLAs, storage demands, and the like. Then, organizations can begin to map their current and future workload plans to the services offered by public cloud providers. As the variety of services offered grows, it is no surprise that enterprises are increasingly adopting multiple cloud providers to match each function to the provider best suited – or with the best price – to handle that function.
If adopting a multi-cloud approach, consideration must be given to where the other pieces of enterprise IT will be located; for example if taking a hybrid approach those database servers, authentication resources, monitoring and management processes all would typically reside on-premises. If taking a multi-cloud but NOT hybrid cloud approach, those resources would have to hosted somewhere – and have access to all enterprise-wide resources to ensure smooth operations.
Gartner suggests the following six step plan when building a cloud strategy, and suggest if you’re not cloud-first at the point you are already behind your competitors.
- Cloud first and multi-cloud: Becoming cloud-first involves the entire organization, not just IT, and involves a holistic approach to evolving business technology to help define the goals and outcomes. Enterprises should be aware that not every application is suitable for the cloud, so cloud-first does not imply abandoning all on-premises workloads or shifting them immediately to cloud providers. Some enterprise applications will never be migrated to public cloud providers until they are completely refactored.
- Continuous placement assessment : Because cloud provider offerings are constantly changing, the best provider for a workload today may not be the best provider for that workload a month from now. Additionally, technology vendors are now offering subscription pricing models that can alter the OpEx/CapEx discussions that have driven many organizations to adopt cloud services in the past. As a result, workloads should be assessed on a regular basis to consider whether they are candidates for migration to a new cloud provider, or even whether they should be moved to a private cloud to leverage attractive vendor pricing.
- Plan for the future : Cloud migration is not a one-time event. Acquiring skills, learning the differences between providers, and developing process improvement plans based on cloud provider capabilities are tasks that must be constantly iterated as applications, on-premises resources, and cloud provider infrastructure evolve. Plan on a multi-year effort for migration, and plan for annual reassessment after that.
- Focus on governance and management : Cloud governance is complex, multi-cloud governance is even more challenging. Because business units can spin up their own services with just a credit card, many organizations do not grasp the entire footprint their IT organizations has within the cloud or clouds. Enterprises that can track consumption of cloud services can begin to manage their multi-cloud, hybrid IT environment as a single entity, or at least ensure that everything under their aegis is properly governed to meet required regulatory demands.
- Find the right tools for the right cloud : Although each cloud provider has their own set of management and deployment tools, organizations can develop a tool strategy that spans multiple clouds by abstracting each provider’s tools layer. For example, Kubernetes can provide cloud orchestration tools that enable applications to be deployed on a single cloud, on a hybrid cloud, or in a hybrid multi-cloud fashion. By investing in a coordinated tool strategy for management and orchestration, enterprises can eliminate many of the headaches that are common in multi-cloud deployments and reduce the time needed to educate IT staff and users alike.
Although the goal is cross-platform consistency, organizations should utilize each cloud platform’s native toolset as well to gain the deepest insights into the arcana of each provider’s platform when needed.
Keep in mind that it is likely that organizations will end up with vendor-specific, cross-platform, and homegrown tools to meet all their management and orchestration needs.
- Consider cloud holistically, including SaaS : Over time, applications hosted on-premises or on an IaaS or PaaS platform may be obsoleted or replaced by less costly SaaS solutions. Each enterprise should be continually evaluating their cloud computing requirements to determine how to fine-tune their overall cloud computing positions, determine if a workload should be hosted elsewhere, or whether to refactor older or legacy applications to take advantage of more agile, cloud-native applications better suited to today’s cloud environments.
The simple answer is yes. A hybrid cloud becomes multi-cloud when there are more than one public cloud service combined with private cloud resources. Hybrid Cloud is a combination of public and private clouds, usually to orchestrate a single IT solution between both. Often a hybrid cloud is built atop a common virtualization layer such as the VMware cloud or vSphere. In this manner, a hybrid cloud that includes VMware on AWS and VMware on-premises simplifies workload migration and enables the use of common tools across both private and public cloud platforms.
Multi-cloud entails multiple cloud services from one or more providers, for example AWS for application workloads and Microsoft Azure for enterprise database. Although it is similar to a hybrid cloud, multi-cloud specifically indicates more than one public cloud provider service and need not include a private cloud component at all (although it can).
Enterprises adopt a multi-cloud strategy so as not to ‘keep all their eggs in a single basket’, for geographic or regulatory governance demands, for business continuity, or to take advantage of features specific to a particular provider.
The main difference between hybrid cloud architecture and multi-cloud architecture is where non-cloud resources are located. Hybrid clouds utilize existing on-premises servers, storage, and networking to support ancillary services such as authentication, VDI, security, databases, and monitoring, wherein a multi-cloud (not hybrid) environment those resources are also in the cloud, either at the same provider providing compute services or another provider or colocation facility.
Although every hybrid cloud can be classified as a multi-cloud by stretching the definition, not every multi-cloud is a hybrid cloud since hybrid is defined as both private and public clouds utilized in concert. For multi-clouds there is no concern about on-premises private cloud infrastructure and attention need only be paid to the individual public cloud services – and ways of simplifying orchestration and monitoring between them. For multi-cloud, admins should focus on a single tool that works across multiple clouds to reduce training, simplify operations, and reduce the opportunity for human errors.
Since there is no private cloud in multi-cloud, organizations must deal with ensuring that where data is stored – whether PCI, HIPAA, or GDPR protected – meets those rigorous demands. Most hyperscale cloud providers offer availability zones and regions to help ensure this, and since even keeping copies or backups of this data outside of the region it is supposed to reside in can cause issues care must be taken when developing a data strategy for the multi-cloud.