Offering speedier innovation, flexible resources, and economies of scale, cloud computing is the supply of computer services such as servers, storage, databases, networking, software, analytics, and intelligence through the Internet (or “the cloud”).

Companies can use cloud computing to gain access to apps and data storage without investing in costly infrastructure or maintaining their own data centres.

Using cloud-computing services allows businesses to pay only for the resources they actually use, rather than the far larger sums required to purchase and maintain their own in-house IT equipment.

Providers of cloud computing services, in turn, can reap substantial cost savings by catering to a large customer base.

Storage, networking, and processing power are just the beginning of what can be done with cloud computing; currently, services extend to more advanced areas like natural language processing, artificial intelligence, and even everyday office software. You can get just about any service, even quantum computing, via the cloud now, as long as you don’t need to be in close proximity to the computer gear you’re utilising.

A plethora of services rely on cloud computing. Services that allow major companies to host all of their data and operate all of their apps in the cloud fall under this category as well as consumer services like Gmail or the cloud backup of your smartphone’s images. For instance, Netflix’s video streaming service and other business operations are entirely dependent on cloud computing.

As more and more software companies move toward a subscription-based business model, cloud computing is quickly becoming the de facto standard for many software solutions. While cloud computing has many benefits, it also has the potential to increase expenses and risk for businesses who adopt it.

Users need not worry about technical particulars such server locations or software versions when using cloud services. For this reason, the cloud metaphor was appropriated from older diagrams of telecommunications networks, in which the public telephone system (and later the internet) was commonly depicted as a cloud to indicate that its physical location was irrelevant and that it was all simply a big blob of data. Of course, this is simplifying things; the location of services and data is still a major concern for many clients.

Although the name “cloud computing” didn’t appear until the early 2000s, the idea of “computing as a service” has been present since the 1960s, when computer bureaus began offering mainframe time for rent rather than outright sale.

The proliferation of personal computers (PCs), which made individual access to computers more affordable, and the subsequent proliferation of corporate data centres (DCs), where businesses could store massive quantities of data, effectively supplanted the need for ‘time-sharing’ services.

However, in the late 1990s and early 2000s, the idea of renting access to computing power reemerged in the form of application service providers, utility computing, and grid computing. The advent of SaaS and large-scale cloud providers like Amazon Web Services paved the way for the widespread adoption of cloud computing.

Computing workloads continue to move to the cloud, whether to public cloud services offered by vendors or to private clouds constructed by organisations themselves, and the cost of constructing the infrastructure to enable cloud computing has become a considerable portion of total IT investment.

Whether we like it or not, the cloud has emerged as the dominant enterprise computing platform.

The market share of cloud-based services is expected to grow from its current 41% to as much as 50% by 2025, according to research from technology analyst Gartner. It forecasts that by 2022, cloud computing would account for 62.3% of all spending on application software, up from 57.7% in 2018.

Many subcomponents of cloud computing exist, each of which addresses a certain facet of the technological stack or a particular set of use cases. In this article, we will examine several of the most well-known in further depth.

Computing’s bare-bones components—physical or virtual servers, storage, and networking—are known as “infrastructure as a service” (IaaS). This is appealing to businesses who prefer to build applications from scratch and have complete control over all aspects of the process, but it does necessitate advanced technical knowledge and expertise to coordinate services at this level.

The next level up is platform as a service (PaaS), which provides not just the storage, networking, and virtual servers but also the middleware, database management, operating systems, and development tools that are required to create applications on top.

When most people think of cloud computing, they typically picture Software as a Service (SaaS), which is the delivery of applications as a service and hence the most common kind of cloud computing. Since the service is typically purchased on a per-seat or per-user basis, the end user should not worry about the underlying hardware or operating system since they will be accessing it via a web browser or mobile app.

Due to the vast number of SaaS-delivered applications, including customer relationship management systems like Salesforce and productivity suites like Microsoft Office 365, SaaS accounts for the lion’s share of cloud spending. According to IDC, while the market as a whole is expanding rapidly, the IaaS and PaaS segments are expanding at much faster rates: “this highlights the increasing reliance of enterprises on a cloud foundation built on cloud infrastructure, software-defined data, compute, and governance solutions as a Service, and cloud-native platforms for application deployment for enterprise IT internal applications.” With “resilience, flexibility, and agility” guiding IT platform selections, IaaS and PaaS are expected to continue growing at a faster rate than the overall cloud industry, as reported by IDC.

While the major cloud providers are eager to meet all of their enterprise customers’ computing needs, many companies are aiming to reduce their reliance on any one provider. The result of this is the proliferation of multi-cloud architectures. This strategy seeks, in part, to discover the optimal combination of technologies across the industry, and in part, to avoid being locked in to just one vendor (which can lead to the sort of high prices and inflexibility that the cloud is often claimed to avoid).

That’s why it’s becoming increasingly important for companies to be able to link and integrate cloud services from a variety of providers. Disadvantages include workflow variations between cloud systems and a lack of staff with experience across different clouds. In addition, customers will want a centralised location from which to manage their various cloud infrastructures, a simplified process for creating and migrating apps and services, and security technologies that are compatible with a wide range of clouds.

The precise advantages will change depending on the specific cloud service, but generally speaking, employing cloud services means that businesses don’t have to invest in and manage their own server hardware.

No more buying servers, updating applications or operating systems, or decommissioning and disposing of hardware or software when it is out of date, as it is all taken care of by the supplier. Making the move to the cloud from in-house expertise might be beneficial for commodity apps like email. Cloud services may be able to provide a more secure and efficient service to end users because they are managed by a company that specialises in running and securing these services and so has access to higher skills and more experienced staff than a small business could afford to recruit.

Because companies only pay for the resources they really use, using cloud services allows them to move more quickly on initiatives and test out concepts without protracted procurement and large upfront expenses. Proponents of cloud computing frequently highlight the concept of corporate agility as one of the main advantages. It should be less of a hassle to get started with new applications quickly thanks to the ability to spin up new services without the time and work associated with conventional IT procurement. Further, the cloud’s elasticity makes it simpler to rapidly scale up a new application if it proves to be really popular.

It may be more cost-effective for a business to host a seasonal or peak-use application on the cloud rather than maintain specialised hardware and software that sits idle for the majority of the year. Moving to a cloud-hosted application for services like email or CRM could relieve pressure on internal IT workers, and if such applications don’t offer any competitive advantage, there will be little else to influence. Some businesses may benefit from shifting from a capex to an opex model, which is what happens when they adopt a services model.

• Business continuity is the ultimate killer application for cloud
• It’s official: Supercomputing is now ho-hum (thanks, cloud)

Just as renting does not always beat out buying, cloud computing may not be less expensive than alternative options. In-house provision of computer services may be more cost-effective if an application has a consistent and predictable demand for such services.

If a service is also used by a company’s competitors, the former may be hesitant to save data that could be used to harm the latter. If a competitor is already utilising the same SaaS applications you are, it could be difficult to gain an advantage over them by switching to a different provider.

While adopting a new cloud service and beginning to use it could be simple, moving already existing data or applications could be a time-consuming and costly process. Staff with expertise in DevOps and multi-cloud monitoring and management appear to be in particularly high demand at the moment.

It was found in one study that many seasoned cloud users believe the one-time expenditures of migrating outweigh the savings made possible by IaaS over the long run.

Of course, you need to be online in order to use any applications at all.

In the cloud, businesses can avoid the upfront costs of purchasing and maintaining their own servers by instead purchasing computing power as a utility. Using the cloud to make room in the budget may be less of a hassle than asking the CFO for more money, allowing businesses to avoid the significant spikes in IT spending typically seen with new initiatives.

However, this does not imply that cloud computing is always or even usually less expensive than keeping applications in-house; in fact, it may be less expensive (at least in terms of processing power) to retain applications in-house if their computing needs are stable and predictable.

• Cloud computing spending is growing even faster than expected

Before making the decision to migrate your systems to the cloud, it is important to get a firm grasp on how much your current infrastructure is costing you. Everything from the obvious, like the price of maintaining data centres, to the ancillary, like leased connections, must be taken into account. Hardware expenses include those for servers and other physical components as well as software and data storage. You’ll also need to figure out how much it’ll cost you for applications, whether you intend to get rid of them, re-host them in the cloud without making any changes, rebuild them from the ground up for the cloud, or switch to a new SaaS provider. The financial repercussions of these alternatives vary. People expenses (typically second only to the infrastructure costs) and more abstract ideas, such the value of being able to supply new services more quickly, must also be factored into the cloud business case. Consequences, such as the possibility of being dependent on a single provider for your IT infrastructure, should be taken into account in any cloud business case (see multi-cloud, above).

• Cloud computing: How to build a business case
• Cloud computing: What it’s like to make the move

Cloud computing is the foundation of many emerging technologies, from mobile banking to healthcare, and its popularity is expected to continue growing, according to industry experts. It’s difficult to imagine many future technological initiatives being completed that don’t use cloud computing in some capacity. According to Gartner, by 2025, more than 85% of businesses will have adopted a cloud-first philosophy because they realise they cannot carry out their digital objectives without it. According to the analyst, new workloads deployed in a cloud-native environment will be ubiquitous rather than trendy, and anything not in the cloud would be deemed old. Over 95% of new digital workloads will be deployed on cloud-native platforms by 2025, according to Gartner, up from 30% in 2021.

And if you think that’s far-fetched, remember that estimates of cloud adoption may vary depending on whom you ask within an organisation. Not all cloud expenditure will be directed by the chief information officer. Since cloud services are simple to set up, business managers can begin making use of them immediately (and paying for them out of their own budget) without informing the IT department. This may help organisations advance more rapidly, but it also poses security dangers if app usage is not regulated.

Different applications will have different rates of adoption; for instance, moving to cloud-based email is significantly simpler than switching to a new financial system. There will also be less immediate need to make a potentially costly and dangerous move to the cloud for systems like supply chain management that are already functioning efficiently.

Although security breaches in the cloud are uncommon, many businesses continue to worry about the safety of using such services. The level of security you expect from cloud computing will depend heavily on the state of your current infrastructure. When compared with the cloud provider’s engineers, who are only focused on the security of their client’s data, in-house systems are more likely to have security holes.

Cloud security solutions, which track information as it travels to and from the cloud, have grown in popularity in response to persistent security concerns, especially among businesses that use many cloud services. Data theft, unauthorised downloads, and malware are all things that can be uncovered with the use of the right technologies in the cloud. However, there is a cost and a performance hit: these technologies can lower cloud ROI by 5-10% and affect performance by 5- 15%. Some businesses are also concerned about the location from which cloud services are provided (for more on this, see “Is geography irrelevant when it comes to cloud computing?” below).

• Cloud security and IoT are the new peanut butter and jelly
• Azure confidential computing: Microsoft boosts security for cloud data
• Three smart cloud services that can help keep your business more secure
• Cloud computing security: This is where you’ll be spending the money
• Security as a Service? We want it, say IT leaders

The traditional concept of cloud computing is the “public cloud,” in which users share a vast amount of computing resources via the internet (whether that is IaaS, PaaS, or SaaS). The capacity to swiftly grow a service is one of the main advantages. The’multi-tenant’ architecture used by cloud-computing providers allows them to distribute their massive amounts of processing power across numerous users. Because of their massive scale, they have enough spare capacity to quickly deal if any particular customer needs more resources; this makes them a popular choice for applications that don’t require as much consistency in their resource needs.

Behind the company’s firewall, private clouds offer many of the same advantages as public ones, but without the security and privacy risks that come with handing over control of sensitive data and applications. For most IaaS or PaaS projects, businesses have complete say over where and how their data is stored, allowing them to provide developers with a secure, elastic, and scalable resource for processing tasks as they arise. However, this heightened level of protection does not come cheap, since most businesses cannot hope to achieve the same economies of scale as the likes of Amazon Web Services (AWS), Microsoft, or Google. However, private cloud could be advantageous for businesses that need extra protection, either to learn about cloud services or to rewrite internal programmes for the cloud before moving them to the public cloud.

Initiating operations on the cloud is a breeze for new businesses. However, this is not the case for the vast majority of businesses, which must instead determine, in light of their existing apps and data, which systems should continue operating as is and which may begin migrating to cloud infrastructure. Migrating to the cloud is a big step that could backfire if organisations fail to account for the full scope of the work involved.

The requirement to rebuild applications to optimise them for the cloud was reported to be one of the largest costs in a survey of 500 early cloud users. This was especially true for more complicated or bespoke applications. One survey respondent in three mentioned expensive data transfer costs as an obstacle to migrating their important applications. Finding personnel with the necessary capabilities for migration is not only expensive, but it also presents a security risk for businesses: if they do locate the proper people, they may be poached by cloud-computing companies with deeper budgets.

In addition, most people still had concerns about the performance of mission-critical applications, with one-third saying that this was the main reason they hadn’t migrated certain mission-critical programmes.

• Cloud computing migration: More expensive and complicated than you thought
• Technology migrations are more painful, and cloud isn’t making them any easier
• Where does the NAS fit in an increasingly cloud-centric world?

It turns out that this is exactly the kind of scenario where cloud computing might be advantageous. Users and providers of cloud computing are being subjected to considerable shifts due to geopolitical factors. First, there’s the problem of latency; if the app is being served from a data centre halfway around the world, or on the other side of a clogged network, you may notice it’s slower than if you were connected locally. That’s the issue with the delay.

The second concern is the question of who owns the information. A lot of businesses, especially in Europe, are understandably concerned about the physical location of the data centres that handle and store their information. Corporations in Europe are concerned that US authorities may gain access to sensitive consumer information if it is held in data centres in the US or (controlled by US companies). So that businesses may keep data close to home, the major cloud service providers have been expanding their regional data centre networks.

To make it even more difficult for authorities in the United States and elsewhere to access consumer data housed in these facilities, some companies have effectively detached some of these datacenters from their core business. Consumers’ information stored in the data centres is managed by a third party entity known as a “data trustee,” and the US government does not have access to it without the consent of customers or the data trustee. To meet the needs of customers who must store data in certain places, cloud service providers will likely establish more data centres in various parts of the world.

Concerns have been raised about cloud security, with the UK government’s cyber security agency advising that government agencies think carefully about the country of origin before integrating cloud services into their supply chains. This was a warning concerning antivirus software, but the problem affects other services as well.

Large datacenters located in various locations across the world power cloud-computing services. Amazon Web Services separates its data centres into “regions” and “availability zones” to better serve its customers. Each AWS region, such as Europe (London) or the United States (Oregon), is further divided into smaller sections called availability zones (AZs). An AZ is made up of many data centres located in such a way that a single disaster cannot bring both of them down at the same time, but still near enough together to support failover in real time for mission-critical systems and other business continuity use cases. AWS offers approximately 80 Availability Zones (AZs), each with multiple connections to the internet and power from several grids.

Similar to how Google divides its cloud-computing resources, each zone contains one or more datacenters from which customers can run their services. There are now eight distinct areas: To safeguard against failures, Google suggests clients deploy apps across many zones and regions.

The Azure cloud computing platform from Microsoft allocates its resources slightly differently. It provides what it calls regions, which are “a collection of data centres located inside a latency-defined boundary and linked together by a regional low-latency network.” Customers with unique data-residency and compliance requirements can take advantage of these “geographies,” which often include two or more locations, “to keep their data and apps close.” In addition, it provides availability zones, which consist of multiple data centres with redundant systems for providing power, cooling, and network connectivity.

Data centres consume a lot of energy; Microsoft, for example, recently signed a 15-year agreement with General Electric to purchase all of the output from GE’s new 37-megawatt wind farm in Ireland. As of 2018, data centres accounted for less than 2% of Ireland’s overall energy usage. By 2026, that number is expected to rise to 15%.

• Cloud computing: IBM overhauls access rules at Euro data centre
• AWS just sold some of its cloud computing infrastructure in China