On-premises	Cloud
SECURITY	IT teams are responsible for the security of all data and workloads.	Providers manage technical administration of security and offer tools for vulnerability management.
RELIABILITY	Even the strongest data center will break at some point. Systems failures and natural disasters can take down entire facilities.	Outages affect availability, so apps that require high availability should stay on premises or be extended across regions for additional redundancy.
COST	Companies pay for the hardware-plus excess capacity if needed-as well as facility operations.	Only pay for what is used. But companies can overpay if a workload isn't properly optimized.
EXPERTISE	Those who have studied and worked with these technologies know the ins and outs of the infrastructure, tape library management, required programming languages and so on.	Takes away the more complex and challenging parts of computing, which allows IT teams to home in on business objectives. It also requires learning skills to adapt to the new environment.

Clouds can be public or private, though public clouds are more commonly associated with cloud computing. Public cloud platforms, such as AWS and Microsoft Azure, pool resources in data centers often distributed around the globe, and users access them via the internet. Resources are provided to customers through metered services, and the cloud vendor is responsible for varying degrees of back-end maintenance.

Private clouds are walled-off environments hosted in a corporate data center or a colocation facility. They lack the massive scale of public clouds. But they do have some elasticity, and a company's developers and administrators can still use self-service portals to access resources. In theory, private clouds provide greater control and security, though it's up to a company's IT team to ensure that happens.

Public clouds and private clouds can be linked to create a hybrid cloud, or two or more public clouds can be connected to create a multi-cloud architecture.

Broadly speaking, there are also three tiers of cloud computing: infrastructure as a service (IaaS), platform as a service (PaaS) and software as a service (SaaS). IaaS pertains to foundational building blocks, such as compute, network and storage. It provides the most flexibility for application development, but it also requires the most overhead. PaaS abstracts those lower-level elements and provides sandbox environments for app developers. The least hands-on cloud model, SaaS, consists of licensed software delivered as web apps.

A public cloud is a platform that uses the standard cloud computing model to make resources -- such as virtual machines, applications, or storage -- available to users remotely. Public cloud services may be free or offered through a variety of subscription or on-demand pricing schemes, including a pay-per-usage model. The main benefits of the public cloud are as follows:

• A reduced need for organizations to invest in and maintain their own on-premises IT resources;
• Scalability to meet workload and user demands;
• Fewer wasted resources because customers only pay for what they use.

Public cloud is an alternative application development approach to traditional on-premises IT architectures. In the basic public cloud computing model, a third-party provider hosts scalable, on-demand IT resources and delivers them to users over a network connection, either over the public internet or a dedicated network.
The public cloud model encompasses many different technologies, capabilities, and features. At its core, however, a public cloud consists of the following key characteristics:

• On-demand computing and self-service provisioning;
• Resource pooling;
• Scalability and rapid elasticity;
• Pay-per use pricing;
• Measured service;
• Resiliency and availability;
• Security and broad network access;

The public cloud provider supplies the infrastructure needed to host and deploy workloads in the cloud. It also offers tools and services to help customers manage cloud applications, such as data storage, security and monitoring.

When selecting a provider, organizations can opt for a large, general-use provider -- such as AWS, Microsoft Azure or Google Cloud Platform (GCP) -- or a smaller provider. General cloud providers offer broad availability and integration options and are desirable for multipurpose cloud needs. Robeeta Cloud providers offer more customization.

Private	Hybrid	Public
A cloud computing model in which an enterprise uses a proprietary architecture and runs cloud servers within its own data center.	A cloud computing model that includes a mix of on-premises, private cloud and third-party public cloud services with orchestration between the two platforms.	A cloud computing model in which a third-party provider makes compute resources available to the general public over the internet. With public cloud enterprises do not have to set up and maintain their own cloud servers in house.
CHARACTERISTICS:	CHARACTERISTICS:	CHARACTERISTICS:
Single-tenant architecture. On-premises hardware. Direct control of underlying cloud infrastructure.	Cloud bursting capabilities. Benefits of both public and private environments.	Multi-tenant architecture. Pay-as-you-go pricing model.

Cloud computing lowers IT operational costs because the cloud provider manages the underlying infrastructure, including hardware and software. Those managed components are typically more reliable and secure than the standard corporate data center. These advantages free IT teams to focus on work that more directly benefits the business.

The cloud is also global, convenient, immensely scalable and easily accessible, all of which accelerate the time to create and deploy software applications. It opens organizations to a host of newer services that enable the most popular trends in application architectures and uses, including microservices, containers, serverless computing, machine learning, large-scale data analytics, IoT and more.

While IT teams lower their Capex with cloud computing because they're not buying gear, they also add significant Opex to their budgets -- often enough to offset most or all their operational savings. Complex pricing and security models can also lead to major problems if IT teams are unable to adapt.

IT teams often must learn new skills or hire employees to navigate the cloud, and there are limits in the flexibility and control over certain cloud resources.

The clouds themselves are generally more secure than most private data centers since companies such as Amazon and Google can hire talented engineers and automate many of their practices. Cloud infrastructure providers also offer tools and architectural options to isolate workloads, encrypt data and detect potential threats.

However, public clouds operate on a shared responsibility model, where the user secures the data and applications hosted on the cloud. This division of security responsibilities varies based on the tier of cloud computing.

The process to secure a cloud environment is different from more traditional data center practices, so cloud adoption requires a learning curve for IT teams. Unauthorized access to resources is the most common cloud security threat; many high-profile exposures of sensitive data resulted from misconfigurations.

Organizations also must be mindful of data residency requirements and other governance restrictions since they don't have precise control over the location of the servers that host their data in the cloud.

In the early days of cloud, most enterprise usage was ad hoc, driven by developers and lines of business that wanted to go around traditional IT procurement processes. Today, organizations must formulate a holistic strategy to successfully move to the cloud. Assemble key stakeholders and employees with cloud experience -- a setup also known as a cloud center for excellence -- to map out a strategy based on business objectives.

Also, a cloud strategy should include a decision framework to identify workload characteristics and how those characteristics port to cloud platforms. IT leaders and cloud architects must evaluate risks and benefits and determine how they will manage and secure cloud-based workloads, as well as whether or how those workloads interact with remaining on-premises assets.

It can take more than a year to execute a cloud migration strategy, and even then, it's an ongoing process. Communication is critical to success. Keep stakeholders regularly informed, and make sure IT is part of the broader decision-making process for the business. Identify leaders who will evangelize the use of the cloud, and ensure employees are properly trained for the transition.

Virtually any workload can move to the cloud, but not all of them should. Large, monolithic applications that run constantly, with relatively consistent and predictable loads, are often more cost-effective on premises. Also, highly regulated enterprises tend to keep applications and sensitive data in their private data centers due to governance and data security concerns.

An organization can move an application as-is to the cloud, but that's typically not the best option. Use a cloud migration to reevaluate your IT architecture and identify greater efficiencies. An application can be broken into microservices to take advantage of the service-based approach of IaaS. Alternatively, a move to PaaS or SaaS can offload undifferentiated heavy lifting.

Public clouds charge on a per-use basis, so costs will vary wildly based on multiple variables, including the size of your environment, the provider, the region you operate in, the amount of data movement and the number of higher-level services consumed.

The major public cloud providers also have pricing schemes that can lower costs in exchange for certain long-term commitments.

There's considerable debate about whether the cloud is ultimately cheaper or more expensive than traditional on-premises computing -- but that misses the point. The cloud can help modernize an enterprise and its IT department and transition IT from its typical role as gatekeeper into a key participant in the decision-making process for the business side of an enterprise.

This depends on the layer of the cloud stack being used. For IaaS, administrators must manage all aspects of the virtualized environment that's delivered as a service. The higher up the cloud stack you go, the less management involved. For SaaS, oversight might be limited to data and identity and access management.

Major cloud providers and third parties offer training and certifications to familiarize IT staff with these processes. Enterprises that plan to transition to the cloud should factor training into every aspect of the process so they're ready to take the reins as soon as the cloud environment goes live.