What is cloud computing?
“The cloud” isn’t a physical place, but rather a method of managing IT resources that largely replaces local machines and private data centers. In the cloud computing model, users access virtual compute, network, and storage resources made available online by a remote provider. Rather than having to buy and maintain extensive computing, storage, and other IT infrastructure—as well has having in-house expertise available to manage this equipment—much of this responsibility is instead taken care of by the cloud services provider.
The Move to the Cloud
Before cloud computing was available, companies had to store all of their data on their own in-house equipment, which they then had to manage and maintain. If a business suddenly needed more capacity due to, for example, a new product launch or an unanticipated increase in orders, they would have to buy and install new hardware, software, and networking infrastructure. This time-consuming process could often limit business opportunities.
Instant Provisioning and Scaling
Today, cloud computing resources can be provisioned instantly over the Internet, with the ability to scale up and down easily as needed. Companies then pay only for the resources they use, as they use them, typically on a monthly basis, without the need to worry about capacity management, hardware refreshes, or other maintenance and management tasks.
Cloud computing models: IAAS, PAAS, and SAAS
Cloud computing can be broken down into three as-a-service (aaS) categories. The most fundamental option for businesses is Infrastructure as a Service (IaaS). With IaaS, users access infrastructure basics such as server space, data storage, and networking, which can be provisioned via an API. System administrators within the business are responsible for managing the company’s databases, applications, security measures, and other factors, while the cloud computing provider manages the servers, hard drives, networking, and storage. This model comes the closest to replicating the functionality of a traditional data center in a hosted environment.
Platform as a Service (PaaS) offers a full-fledged development environment, eliminating the need for developers to deal directly with the infrastructure layer when deploying or updating applications. In addition to the elements of IaaS, PaaS includes various tools and software that developers need to build, test, and run their applications. PaaS allows developers to focus on creative tasks without worrying about managing operating systems, databases, middleware, and development tools.
Software as a Service (SaaS) is the category the majority of users are likely most familiar with. SaaS applications are designed directly for end users, with the underlying infrastructure remaining behind the scenes. Users typically access the service via a web browser or app, and billing is often per user or per seat. From business applications such as word processing and spreadsheet programs, to customer relationship management (CRM) and enterprise resource management (ERM) applications, to photo-editing suites and video-hosting platforms, SaaS apps provide a huge range of functionality in the cloud.
Cloud computing deployment options: Public, private, and hybrid
Cloud deployment models are defined by where the cloud infrastructure is physically located, who has control of the infrastructure, and how cloud services are made available to users. Three cloud computing models are most relevant in the enterprise space, and each meets different business needs.
Public cloud is the classic model, in which shared cloud infrastructure is owned, maintained, and managed by a cloud service provider (CSP). Services are delivered over the open Internet. The largest cloud service providers with data centers that enable massive scaling are called hyperscalers; examples include Amazon Web Services (AWS), Google Cloud Platform (GCP), and Microsoft Azure. The chief benefits of public cloud are its on-demand scalability and pay-as-you-go pricing. Common uses for public cloud computing include file sharing, email service, and application development and testing.
A private cloud runs behind a firewall on an enterprise’s intranet and is hosted in an on-premises or colocated data center. In this model, infrastructure is used by a single organization and can be configured and managed according to that company’s specific needs; data stored in a private cloud remains fully under the company’s control. Private clouds are often used for mission-critical, secure applications, as this model best addresses many organizations’ privacy and security concerns.
Hybrid cloud allows companies to leverage both public and private cloud computing solutions, including the use of multiple public cloud providers. With hybrid cloud, organizations can take advantage of the strengths of each cloud model. It also enables flexibility and scalability, protects sensitive data and operations, and minimizes reliance on any single CSP. A significant benefit of hybrid cloud systems is that they do not have a single point of failure, and they are an attractive option for companies dealing with highly variable workloads.
Common uses for cloud computing
Enterprises use cloud computing for a variety of purposes, depending on their business environment and needs. But several applications tend to be the most commonly used by organizations who have moved at least some of their apps and data to the cloud.
Data storage: As data volumes grow, traditional storage infrastructure models lack the agility to meet rapidly changing capacity demands. Not only does the cloud model make it easy for companies to gain new capacity when needed, the cloud service provider typically manages and maintains the storage infrastructure, sparing companies from having to oversee these routine IT tasks.
Data protection: With unpredictable data growth, capacity planning for backup and recovery is often slow and complex. Cloud computing offers a modern solution for the challenges of backing up and protecting critical business data. Cloud providers typically have facilities in many different geographic locations, mitigating the risk of disruption due to natural or man-made disasters, and have robust physical security at those locations.
Big Data analytics: Tools such as Hadoop and Splunk make it easier to ingest and analyze massive amounts of data—but that data has to live somewhere. Cloud computing gives enterprises the option of storing large amounts of data on servers they do not have to physically manage themselves.
Private or hybrid cloud: Enterprises that want the flexibility of the public cloud but need to maintain ownership of their data can adopt a private cloud approach with virtualized resources, which avoids handing over control of their data to a third-party provider. In a hybrid cloud environment, the company uses both private and public cloud resources, with some resources offloaded to a third party while others are kept on premises for security, privacy, or regulatory reasons.
Application testing and deployment: Cloud-based virtualized infrastructure makes it easier for DevOps teams to access resources specifically designed to support a wide range of applications. Containers can be set up to run cloud-native, non-cloud-native, and legacy apps, and container management tools such as Kubernetes streamline deployment and management.
HPE GreenLake cloud services
Many companies value the inherent flexibility and ease of use of the cloud computing experience. However, 70 percent of enterprise apps and data still remain outside the public cloud. 1 Issues such as data gravity, compliance, app dependency, performance, and security require some apps and data to remain hosted in colocations, data centers, and, increasingly, at the edge. HPE GreenLake brings the cloud experience to your apps and data wherever they live and delivers visibility and control across all your clouds, in a single operating model.
The market-leading HPE GreenLake cloud services portfolio features modular building blocks that enable workloads with a stack of infrastructure, software, and services. This pre-configured, workload-optimized hardware and software can be delivered in as few as 14 days to your owned or colocated data center facility. Solutions are available for a variety of workloads, such as:
- Data protection
- Big Data
- Private cloud
- Machine learning
- Virtual machines
- High performance computing (HPC)
- Virtual desktop infrastructure (VDI)
Migrating to the hybrid cloud—with its combination of on-premises, edge, and public cloud resources—is a complex and lengthy process. It requires you to determine the right mix of destination choices for your business applications and enables you to execute a hybrid cloud migration plan. HPE Right Mix Advisor provides an objective, data-driven analysis that prepares your business for successful hybrid cloud migration. The service leverages HPE’s experience and insights gained from many successful enterprise application migration engagements.
HPE also delivers services to manage your end-to-end hybrid cloud environment. These services take the management burden from you, giving you the ability to access, consume, monitor, and control all your on- and off-premises cloud services and infrastructure from a single client platform, no matter the vendor. Our award-winning management services utilize an advanced suite of integrated tools, IP, processes, and best practices to manage and optimize your entire hybrid cloud environment, driving greater time to value and reducing costs and risk. 2