The cloud is clearly the present and the future of computing and technology infrastructure. Organizations of all sizes, in pretty much every vertical, have already adopted and deployed cloud-based apps and services as part of their current IT functions. Whether it’s public, private or hybrid cloud, these technologies will be absolutely crucial to building an exciting and robust future in all things robotics, drones, sensors, artificial intelligence and machine learning.
Cloud has evolved to encompass public (or off-premises) cloud services and private (or on-premises) cloud technology-powered data centers.
Artificial intelligence and machine learning in particular has driven an explosion in cloud needs. The incredibly vast amounts of data required to train machine learning algorithms demands the amazing computing power and flexibility of the cloud. With generative AI tools like ChatGPT gaining traction as well, cloud computing will continue its exponential climb in growth.
Public cloud is one cloud delivery model that continues to grow while cloudification of owned data centers is becoming the norm with virtualization and hyperconverged infrastructure (HCI) technologies. With the technology and cloud markets having matured significantly, IT leaders have clearly indicated that business outcomes take precedence over cutting costs. That’s why 90% of respondents are taking a “cloud smart” approach to their infrastructure strategy, according to the 2024 Enterprise Cloud Index. The hybrid cloud infrastructure remains a draw because it offers a great many business benefits over competing IT deployment models.
Public cloud is one growing sector with both single and multiple public cloud use expected to double over the next 1-3 years according to respondents plans. This indicates that businesses are looking for the flexibility delivered by different public cloud providers.
“The cloud’s value proposition has been validated,” said Sid Nag, VP Analyst at Gartner. “The ability to use on-demand, scalable cloud models to achieve cost efficiency and business continuity is providing the impetus for organizations to rapidly accelerate their digital business transformation plans. The increased use of public cloud services has reinforced cloud adoption to be the ‘new normal,’ now more than ever.”
While the business benefits are clear, how do organizations choose which functions or workloads should run in private or public clouds? For this, they need to understand the different types of cloud services, how they satisfy a specific set of business needs and how they can be moved between private and public cloud if and when it makes sense.
What are Cloud Services?
Cloud services give the organization different levels of IT infrastructure capabilities encapsulated under a software framework. They are “delivered” by vendors or providers to consumers or end users by way of three major models:
Software as a Service (SaaS): The vendor builds and manages applications for various business functions and delivers them to users via the internet. These apps are hosted on the provider’s infrastructure. On the client side, they run on web browsers and don’t require any downloads or installations.
Platform as a Service (PaaS): Developers on the client side can create customized applications, and deploy and manage them on the provider’s infrastructure. The provider manages the underlying OS, middleware, and hardware virtualization.
Infrastructure as a Service (IaaS): The consumer has the freedom to architect, deploy, and manage the application on the virtual resources provided by the vendor. They can buy and automate VMs and virtualized hardware resources on-demand instead of having to purchase hardware while the provider manages the actual servers, storage, and networking.
Artificial Intelligence as a Service (AlaaS): Companies can utilize a wide swath of capabilities and tools powered by AI without needing to invest in AI specific personnel or hardware themselves. These providers often host algorithms for machine learning.
Compared to on-premises datacenters where IT needs to purchase, manage, and upgrade all software and hardware resources, cloud services help reduce the twin burdens of skilling up and CAPEX for technology.
Apart from these, there are secondary -as-a-Service cloud offerings that give more control over particular workloads to customers. These services are highly customized and function-specific. More on this later.
A distinction of note: cloud service models are different from cloud deployment models. The deployment model – such as public, private, or hybrid cloud – defines the underlying architecture over which cloud services are implemented or delivered.
Now for a closer look at the primary cloud service models…
Software as a Service (SaaS)
SaaS offers the whole package from applications to OS to infrastructure that users can seamlessly access with the simplest of devices and minimal bandwidth. The amount of software services available is vast and caters to every imaginable business function, such as invoicing, project management, file sharing, communications, and so on.
Examples: Dropbox, Slack, Basecamp
The Benefits of Cloud Services:
Lightweight: SaaS is the ideal choice for small businesses that can do with off-the-shelf software and don’t have the in-house IT skills or resources to deploy on-premises hardware.
Affordable: Companies no longer have to purchase facilities and acquire vast quantities of hardware to build their own datacenter. When employing cloud services, there are no upfront infrastructure setup costs. Software licensing costs are typically charged on a pay-per-use subscription model.
Scalable: Businesses can easily scale operations up and down as circumstances change and evolve without having to deal with large investments in equipment.
Highly accessible: SaaS services can be set up and configured in no time. Further, they can be accessed using virtually any device – from a smartphone to a bulky desktop – from any place with an internet connection.
Limitations and Concerns:
Vendor lock-in: SaaS providers make it easy to get into but not so easy to get out of their ecosystem. This limits integration and interoperability with other apps and services. Data portability is limited by the file formats or standard APIs that the vendor provides.
Limited features: One-size-fits-all is the basic premise of SaaS apps, so there is very limited scope for customization. Customers are limited to out-of-the-box features.
Lack of control: Data security and governance lies in the hands of the provider. Customers have little control over the software in terms of appearance, updates, or data storage and transfer methods. This might lead to compromised security, inadequate performance, or downtime.
Platform as a Service (PaaS)
PaaS provides special software components, APIs, scripts, and tools that allow consumers to build scalable and highly available applications that conform to popular architectures. This lets developers quickly build, test, and deploy apps iteratively without the need for extensive internal infrastructure resources. These apps might not be fully customizable but the shortening of the development cycle and simplicity of management more than compensate for that.
Examples: Heroku, Cloud Foundry, AWS Elastic Beanstalk
Benefits:
Fast development: With on-demand availability of server-side components and compute resources, developers can speed up testing and production processes.
Multiple language support: Inherent support for all major programming languages gives developers the flexibility to build software that delivers the highest performance for any given workload or project.
Collaboration: Development teams spread over different geographic locations can develop and host applications in the same environment without the need for transferring or syncing data, or files. Everyone has access to up-to-date information and code versions in real-time.
Limitations and Concerns:
Data security: Even though companies run their own apps and services, ultimately the data resides in the vendors’ servers, raising security and compliance concerns.
Complexity in integration: The provider may not necessarily have the integration modules to connect legacy apps and services to the cloud. This might warrant extra customization on the client side.
Infrastructure as a Service (IaaS)
IaaS gives enterprises the ability to build and use a “datacenter in the cloud” without buying expensive hardware outright. When a company needs to rapidly deploy IT infrastructure but doesn’t have the ability to do so in-house, IaaS provides on-demand access to server, storage, or networking resources via the internet. IaaS also offers the choice of a public, private, or hybrid cloud environment.
Examples: Digital Ocean, Linode, Rackspace, Microsoft Azure
Benefits:
Flexibility: Organizations can scale up (or down) the infrastructure and VMs on demand, according to project-specific requirements.
Reliability: The underlying hardware is housed across remote datacenters, eliminating the threat of a single point of failure. Further, the infrastructure is virtualized, provisioned, and managed by skilled IT personnel at major providers, minimizing the possibility of downtime.
Limitations and Concerns:
Training for staff: The organization needs resources and training for IT admins to be able to effectively monitor and manage the infrastructure.
Security: IaaS uses a multi-tenancy architecture for storage and server resources. The onus is on the vendor to prevent unauthorized access to customer data on storage assets or VMs.
Everything as a Service (XaaS)
XaaS is an umbrella term that encompasses any service that delivers a specific function or operational facility over the cloud. Here are a few examples:
Data as a Service (DaaS): The provider delivers cloud-hosted virtual desktops over the internet, licensed on a per-user basis.
Database as a Service (DBaaS): These services enable the setup, configuration, operation, and scaling of databases without the need for physical servers or storage on the client side.
Desktop as a Service (DaaS): A range of data services, such as analytics, storage, data warehousing, data processing, and data virtualization that are provided on-demand via the cloud, along with software tools for managing and working with data.
Unified Communication as a Service (UCaaS): UCaaS is a cloud delivery mechanism for enterprise communications. It integrates phone, email, instant messaging, and video conferencing into an integrated system.
Business Process as a Service (BPaaS): Strategically combining elements of SaaS and PaaS, BPaaS packages process design, implementation, and optimization capabilities into a software that can be customized by clients with their own logic.
Security as a Service (SECaaS): The provider packages data and information security services – including intrusion detection, penetration testing, and antivirus – into a subscription that reduces the TCO of security for client organizations.
Choosing a Cloud Delivery Model
Organizations need to carefully evaluate which model and architectures will work best for their business needs, objectives, and projects.
For instance, think about what level of control does your business require? What kind of technical expertise does your team have? What are the various security and scalability needs? How about the company’s budgetary demands?
Each cloud service delivery model has its own set of pros and cons and as a whole, they cover a wide variety of potential requirements, which is why companies must mix cloud usage. To mix clouds most cost-effectively, workloads need to be strategically allocated to providers depending on their pricing differences and every IT need must be constantly monitored for the best optimization.
“Continually re-evaluate your use of the cloud to see if there are new cloud services or features that you should leverage,” said Gokul Rajagopalan, Director of Products at Vectra AI. “These optimizations can result in significant improvements in performance or reductions in costs that are easy to miss.”
Meanwhile, cloud technology keeps influencing change and disruption across all business segments, even more so as technologies like artificial intelligence and machine learning gain widespread adoption. It’s safe to claim that innovation and experimentation driven by cloud services will continue to bolster organizational capabilities.
In such a scenario, Gartner forecasts worldwide spending on public cloud services to top $723 billion by the end of 2025.
“Cloud will serve as the glue between many other technologies that CIOs want to use more of, allowing them to leapfrog into the next century as they address more complex and emerging use cases,” said Nag.
Chase Guttman updated this article, wihch originally published on September 22, 2021.
Dipti Parmar is a marketing consultant and contributing writer to Nutanix. She writes columns on major tech and business publications such as IDG’s CIO.com, Adobe’s CMO.com, Entrepreneur Mag, and Inc. Follow her on Twitter @dipTparmar and connect with her on LinkedIn.
Featured image by Pxfuel
© 2024 Nutanix, Inc. All rights reserved. For additional information and important legal disclaimers, please go here.
