What is cloud computing?

Cloud computing, sometimes referred to simply as “cloud,” is the use of computing resources — servers, database management, data storage, networking, software applications, and special capabilities such as blockchain and artificial intelligence (AI) — over the internet, as opposed to owning and operating those resources yourself, on premises. 

Compared to traditional IT, cloud computing offers organizations a host of benefits: the cost-effectiveness of paying for only the resources you use; faster time to market for mission-critical applications and services; the ability to scale easily, affordably and — with the right cloud provider — globally; and much more (see “What are the benefits of cloud computing?” below). And many organizations are seeing additional benefits from combining public cloud services purchased from a cloud services provider with private cloud infrastructure they operate themselves to deliver sensitive applications or data to customers, partners and employees.

Increasingly, “cloud computing” is becoming synonymous with “computing.” For example, in a 2019 survey of nearly 800 companies, 94% were using some form of cloud computing (link resides outside IBM). Many businesses are still in the first stages of their cloud journey, having migrated or deployed about 20% of their applications to the cloud, and are working out the unique security, compliance and geographic implications of moving their remaining mission-critical applications. But move they will: Industry analyst Gartner predicts that more than half of companies using cloud today will move to an all-cloud infrastructure by next year (2021) (link resides outside IBM).

A brief history of cloud computing

Cloud computing dates back to the 1950s, and over the years, it has evolved through many phases that were first pioneered by IBM, including grid, utility and on-demand computing.

To read a full history of cloud computing, from mainframes to how virtualization introduced the modern-day cloud, check out IBM’s history of cloud computing blog post.

What are the benefits of cloud computing?

Compared to traditional IT, cloud computing typically enables:

  • Greater cost efficiency. While traditional IT requires you to purchase computing capacity in anticipation of growth or surges in traffic — capacity that sits unused until you grow or traffic surges — cloud computing enables you to pay for only the capacity you need, when you need it. Cloud also eliminates the ongoing expense of purchasing, housing, maintaining and managing infrastructure on premises.
  • Improved agility; faster time to market. On the cloud you can provision and deploy  (“spin up”)  a server in minutes; purchasing and deploying the same server on premises might take weeks or months.
  • Greater scalability and elasticity. Cloud computing lets you scale workloads automatically — up or down — in response to business growth or surges in traffic. And working with a cloud provider that has data centers spread around the world enables you to scale up or down globally on demand, without sacrificing performance.
  • Improved reliability and business continuity. Because most cloud providers have redundancy built into their global networks, data backup and disaster recovery are typically much easier and less expensive to implement effectively in the cloud than on premises. Providers who offer packaged disaster recovery solutions— referred to disaster recovery as a service, or DRaaS — make the process even easier, more affordable and less disruptive.
  • Continually improving performance. The leading cloud service providers regularly update their infrastructure with the latest, highest-performing computing, storage and networking hardware.
  • Better security, built in. Traditionally, security concerns have been the leading obstacle for organizations considering cloud adoption. But in response to demand, the security offered by cloud service providers is steadily outstripping on-premises solutions. According to security software provider McAfee, today 52% of companies experience better security in the cloud than on premises (link resides outside IBM). Gartner has predicted that by this year (2020), infrastructure as a service (IaaS) cloud workloads will experience 60% fewer security incidents than those in traditional data centers (link resides outside IBM).

With the right provider, cloud also offers the added benefit of greater choice and flexibility. Specifically, a cloud provider that supports open standards and a hybrid multicloud implementation (see “Multicloud and Hybrid Multicloud” below) gives you the choice and flexibility to combine cloud and on-premises resources from unlimited vendors into a single, optimized, seamlessly integrated infrastructure you can manage from a single point of control — an infrastructure in which each workload runs in the best possible location based on its specific performance, security, regulatory compliance and cost requirements.

Cloud services

IaaS – Infrastructure as a Service

The original cloud computing service, IaaS provides foundational computing resources — physical or virtual servers, operating system software, storage, networking infrastructure, data center space — that you use over an internet connection on a pay-as-you-use basis. IaaS lets you rent physical IT infrastructure for building your own remote data center on the cloud, instead of building a data center on premises. IaaS remains the fastest-growing cloud services segment; according to Gartner, it will grow 24% this year.

PaaS – Platform as a Service

PaaS provides a complete cloud-based platform for developing, running and managing applications without the cost, complexity and inflexibility of building and maintaining that platform on premises. The PaaS provider hosts everything — servers, networks, storage, operating system software, databases — at its data center. Development teams can use all of it for a monthly fee based on usage, and can purchase more resources on demand, as needed.

With PaaS you can deploy web and mobile applications to the cloud in minutes, and innovate faster and more cost-effectively in response to market opportunities and competitive threats.

PaaS is one of the fastest-growing cloud offerings today; Gartner forecasts the total market for PaaS to exceed $34 billion by 2022 (link resides outside IBM), doubling its 2018 size.

Serverless Computing

Serverless computing (typically referred to simply as “serverless”) is a hyper-efficient PaaS, differing from conventional PaaS in two important ways:

  • Serverless offloads all responsibility for infrastructure management tasks (scaling, scheduling, patching, provisioning)  to the cloud provider, allowing developers to focus all their time and energy on code.
  • Serverless runs code only on demand — that is, when requested by the application,  enabling the cloud customer to pay for compute resources only when they code is running. With serverless, you never pay for idle computing capacity.

To learn more about serverless — including which workloads are prime targets for serverless, and which are not — read this post.

SaaS – Software as a Service

SaaS is application software that runs in the cloud, and which customers use via internet connection, usually in a web browser, typically for a monthly or annual fee. SaaS is still the most widely used form of cloud computing. If you’ve used salesforce.com, Hubspot or Carbonite, you’ve used SaaS.

SaaS lets you and your team start using software rapidly. Just sign up and get to work. It lets you access your specific instance of the application and your data from any computer, and typically from any mobile device. If your computer or mobile device breaks, you don’t lose your data (because it’s all in the cloud). The software scales as needed. And the SaaS vendor applies fixes and updates without any effort on your part.

Types of cloud computing

Generally speaking, there are four cloud computing models.

Public cloud

In a public cloud, a cloud provider offers affordable access to cloud services, running on some portion of its privately-owned infrastructure, via the internet. Customers don’t need to purchase any hardware, software or supporting infrastructure of their own; everything is owned and managed by the cloud provider, and the customer effectively “rents” a portion of it for a subscription or usage-based fee. Multiple customers — thousands and thousands — share the resources of a public cloud.

Public cloud offers instant access to SaaS business applications, IaaS computing and storage resources, and PaaS for application deployment and development. According to Gartner, customers will spend $266.4 billion on public cloud services this year (link resides outside IBM), up 17% from last year.

Private cloud

Private cloud is cloud infrastructure operated exclusively for one company; it’s managed by the company or a third party (or both), and is hosted primarily on premises, but can also be hosted on dedicated cloud-provider or third-party infrastructure. Private cloud enables a company to take advantage of cloud efficiencies while providing greater control over resources, data security and regulatory compliance, and avoiding the potential impact of sharing resources with another cloud customer (called “shared tenancy”).

Hybrid cloud

Hybrid cloud integrates private and public clouds, using technologies and management tools that allow workloads to move seamlessly between both as needed for optimal performance, security, compliance and cost-effectiveness. For example, hybrid cloud enables a company to keep sensitive data and mission-critical legacy applications (which can’t easily be migrated to the cloud) on premises, while leveraging public cloud for SaaS applications, PaaS for rapid deployment of new applications, and IaaS for additional storage or compute capacity on demand.

Multicloud and hybrid multicloud

Multicloud refers to infrastructure comprising multiple vendors’ public clouds — the use of services from two or more major cloud providers (e.g., IBM Cloud and Google), or services from a major cloud provider and at least one SaaS software vendor.

Hybrid multicloud refers to the use of private cloud plus multicloud.

Increasingly, businesses are embracing hybrid multicloud as the deployment model that lets them take maximum advantage of cloud agility and flexibility; meet their security and regulatory compliance needs; and integrate their legacy applications and systems. According to industry analyst 451 Research (Video, 02:03), at least 67% of businesses use two or more separate clouds, and 55% or organizations have settled on a hybrid multicloud approach. IBM estimates hybrid cloud to be a $1.2-trillion business opportunity for cloud and cloud service providers.

Cloud computing and IBM Cloud

IBM Cloud™ is a robust suite of advanced data and AI tools, and deep industry expertise to help you on your journey to the cloud. Customers can choose among over 170 products and services covering data, containers, AI and machine learning, IoT, blockchain and more. And they can combine public cloud, public dedicated cloud, private cloud, hybrid cloud and multicloud deployment models to match the right workload to the right cloud environment.

IBM Cloud maintains 60 data centers worldwide, enabling local deployment, global scalability, and built-in resiliency and redundancy on six continents. 47 of the Fortune 50 companies trust mission-critical applications to IBM Cloud’s enterprise-grade infrastructure. And IBM is the one cloud provider that’s also a leading IT security organization, with over 8,000 security experts on staff.

Learn how American Airlines became more responsive to customer needs with a new technology platform from the IBM Cloud.

To get started with IBM Cloud, sign up for an IBMid and create your IBM Cloud account.

Security issues in cloud computing

Your cloud infrastructure is only as secure as you make it. Responsibility for securing the cloud lies not only with security teams, but also with DevOps and operations teams that are charged with ensuring appropriate security controls are used. Businesses are eager to bring more regulated workloads to the cloud, including any application that manages or contains personal identifying information, financial information or healthcare information.

To avoid cloud computing risks, a cloud managed services provider should incorporate built-in security layers at every level — from the data center to the operating system — delivering a fully configured solution with industry-leading physical security and regular vulnerability scans performed by highly skilled specialists.

Cloud computing for the enterprise

Enterprises eager to undergo digital transformations and modernize their applications are quick to see the value of adopting a cloud computing platform. They are increasingly finding business agility or cost savings by renting software. Each cloud computing service and deployment model type provides you with different levels of control, flexibility and management. Therefore, it’s important to understand the differences between them.

Common convention points to public cloud as the delivery model of choice. But, when considering the right architecture of cloud computing for your applications and workloads, you must begin by addressing the unique needs of your business.

This can include many factors, such as government regulations, security, performance, data residency, service levels, time to market, architecture complexity, skills and preventing vendor lock-in. Add in the need to incorporate the emerging technologies, and you can see why IT leaders are challenging the notion that cloud computing migration is easy.

At first glance, the types of cloud computing seem simple: public, private or a hybrid mix of both. In reality, the choices are many. Public cloud can include shared, dedicated and bare metal delivery models. Fully and partially managed clouds are also options. And, in some cases, especially for existing applications where architectures are too complex to move or the cost-benefit ratio is not optimal, cloud may not be the right choice.

The right model depends on your workload. You should understand the advantages and disadvantages of each cloud deployment model and take a methodical approach to determining which workloads to move to which type of cloud for the maximum benefit.

Dive deeper into specific cloud service and deployment models, cloud computing architecture and cloud computing examples

Cloud computing storage

Storage growth continues at a significant rate, driven by new workloads like analytics, video and mobile applications. While storage demand is increasing, most IT organizations are under continued pressure to lower the cost of their IT infrastructure through the use of shared cloud computing resources. It’s vital for software designers and solution architects to match the specific requirements of their workloads to the appropriate storage solution or, in many enterprise cases, a mix.

One of the biggest advantages of cloud storage is flexibility. A company that has your data or data you want will be able to manage, analyze, add to and transfer it all from a single dashboard — something impossible to do today on storage hardware that sits alone in a data center.

The other major benefit of storage software is that it can access and analyze any kind of data wherever it lives, no matter the hardware, platform or format. So, from mobile devices linked to your bank to servers full of unstructured social media information, data can be understood via the cloud.

Cloud computing pricing

Like any other business-critical decision, selecting a cloud service requires due diligence and research beyond published per-unit rates. It requires an in-depth understanding of workload performance characteristics and needs and the ability to match those needs to the actual offerings of multiple cloud vendors.

Cloud service providers have no shared standard “units” for cloud capacity or common pricing structures, nor are there common specifications for the underlying hardware that runs the cloud applications. As a result, an assumption of total workload cost based on a provider’s basic “per unit” rate can easily be off by orders of magnitude.

10 tips for enterprise cloud procurement

  • Do your homework. Don’t assume that the provider that’s currently in the news with price decreases will be the best-priced provider for your workload.
  • Understand all workload requirements that will impact cloud workload costs and operations (not just compute and storage). Consider costs associated with licensing software for each core, data transfer to the internet or private network and persistent storage.
  • Understand how the provider will support geographically-dispersed workloads. If your application must move data throughout the globe, ensure that the provider not only has data centers in the regions where you do business, but also a high-performance, private global network. Also consider whether the provider charges data transfer fees between and among cloud centers — any such fees can considerably add to costs if your company expands globally.
  • Consider your business requirements, including the “agility” tradeoff. Can you afford to lock in to a specific provider, unit type, volume or time frame, even if it means a discount?
  • Consider the “net present value of money” when you evaluate long-term pricing options. Seek input from your finance department, especially if you are considering an upfront payment option. This will ensure your comparisons are valid and adhere to your company’s accounting rules for net present value.
  • Factor in the non-workload-specific costs your business will need to run the workload optimally, including technical support, engineering and even professional services.
  • Allow for changing workload needs. You should be able to move workloads as needed from, for example, bare metal to virtualized servers without a major effort.
  • Consider the big picture. Each cloud workload should fit into a holistic cloud strategy, one that will likely comprise multiple deployment models, geographies and vendors.
  • Even as you consider price performance based on the individual cloud workload, consider the provider’s ability to support your broader hybrid IT strategy via OpenStack-compatible platforms, integrated solutions and seamless migration across models.
  • Finally, don’t just look at price for the workload. Instead, consider price-performance to be your base unit of comparison as you consider cloud options or any type of IT option.

Use the cloud pricing calculator to get an idea of your workload costs

The future of cloud

Within the next three years, 75 percent of existing non-cloud apps will move to the cloud. Today’s computing landscape shows companies not only adopting cloud but using more than one cloud environment. Even then, the cloud journey for many has only just begun, moving beyond low-end infrastructure as a service to establish higher business value.