What Is PaaS? How Platform as a Service is Different from IaaS and SaaS

PaaS (Platform as a Service) combines servers, storage, and network infrastructure with the software you need to deploy apps. With PaaS, you don’t need to invest in on-premise hardware or worry about setting up a virtual environment to handle your apps.

Almost 85% of organizations were expected to have most of their workloads in “the cloud” by the end of 2020. These companies can then use cloud-based data to personalize marketing and share strategic data between departments (avoiding silos)—two main digital transformation goals.

PaaS gives you all the tools you need to realize these goals and more.

Let’s explore what PaaS is, some real-world PaaS examples, the market share of leading PaaS providers, and more.

What Is PaaS? (With Examples)

Platform as a Service (PaaS) refers to a pre-packaged combination of cloud computing hardware and software tools that let you develop and deploy applications with ease.

For example, AWS Elastic Beanstalk is a plug-and-play platform that supports multiple programming languages and environments straight out of the box. It makes it easy to deploy and quickly test, launch, and scale apps for different devices across several platforms.

It automatically deploys uploaded code and includes load-balancing and auto-scaling tools to keep your apps running smoothly at all times.

Google App Engine is another example of a PaaS for software deployment.

Beyond deploying an app in any language, you also get app diagnostics, versioning, A/B testing, and other features.

These services are fundamentally different from just renting storage space or a virtual machine. A PaaS solution serves as a ready-made platform to take your apps live.

Renting the hardware alone falls under IaaS, not PaaS.

Let’s take a closer look at the differences between these two categories.

What Is the Difference Between IaaS and PaaS?

Infrastructure as a Service (IaaS) refers to cloud services’ hardware components, like virtual machines and scalable storage. PaaS also includes an ecosystem for deploying your applications.

They’re both examples of cloud computing services, of course, but with different use cases and target markets.

To make things even more confusing, many PaaS providers also offer IaaS solutions and vice versa.

Let’s clear up the confusion by comparing specific products.

For example, using an AWS EC2 instance to store data for your web app means you’re only taking advantage of fundamental cloud infrastructure. So, in this case, you’re only using an IaaS service.

If you use AWS Elastic Beanstalk to deploy a web application instead, you’d be using a PaaS service. The Elastic Beanstalk implementation uses AWS infrastructure like S3, EC2, and DynamoDB, but combines them into an instantly usable platform for development.

That’s the difference.

A PaaS is a packaged solution ready to help you develop and deploy your app, while an IaaS is just the bare-bones cloud infrastructure.

It’s similar to the difference between an unmanaged web hosting service (where you have to install the server environment like PHP & MySQL) and an application, database, and managed WordPress hosting platform (like Kinsta).

In some cases, vendors like AWS, Google, and Microsoft will only charge you for resource usage and nothing for using the software platform.

What Is the Difference Between SaaS and PaaS?

Software as a Service (SaaS), on the other hand, offers complete software, not a platform you can use to deploy custom applications.

For example, a cloud-based CRM like Salesforce lets you manage customer data and workflows without any programming.

All you need to do is sign up, and it’s ready to use. That’s the principle of a SaaS service.

In comparison, to do the same with a PaaS, you’d need to develop a custom app or deploy and customize an open source software with similar functionality.

To understand the difference, let’s examine some more examples.

A SaaS product is ready to use the second anyone signs up. For example, once you join Google Docs, you can start creating, saving, and sharing documents.

But Google App Engine or IBM Cloud Foundry aren’t finished applications that solve business problems. Instead, they’re cloud platforms on which you can deploy custom applications.

Most companies use a combination of SaaS (we use 40+ SaaS products) and PaaS to run their business. The optimal solution depends on each use case and your company’s experience and infrastructure.

PaaS Market Size, Share, and Leading Vendors

The PaaS market’s reported size and how it compares to other cloud services depend on the source.

For example, according to Gartner, PaaS will be dwarfed by IaaS in 2021, with $27.5 billion vs. $61.9 billion in revenue, respectively.

That would make the IaaS market more than twice the size of PaaS. IaaS is a significant part of the entire cloud services market, while PaaS seems almost niche in comparison.

But other research shows a different picture.

According to International Data Corporation (IDC) data, 2019 revenues were a lot closer, and PaaS a lot larger, at $35.9 billion versus $49 billion.

Both categories are growing at almost the same rate, with 38.4% YoY growth for IaaS and 38.8% growth for PaaS.

While PaaS might not be closing the gap, it’s not falling behind either.

The cloud market share between the different cloud services categories depends on defining the borders between them. Notice that IDC doesn’t single out management and security services or business process services in its breakdown.

That may be the reason SaaS and PaaS are significantly larger in its table.

Leading Vendors and Their Market Share

Since 2016, cloud industry pundits, research firms, and experts have transitioned from separate reporting to covering the public cloud industries of IaaS and PaaS combined.

One of the main reasons for this is that the leading players are mostly the same across both categories.

You’ll see familiar names like Amazon, Google, Microsoft, and IBM, whether you’re analyzing the IaaS or PaaS markets.

The second reason is that these market leaders tend to bundle both IaaS and PaaS services together, so it can be hard to separate the revenue.

For example, AWS gets a lot of its IaaS revenue because it offers some PaaS environments (like Elastic Beanstalk) for free. Since people only pay for resource usage, it’s hard to single out how much revenue comes from the platform versus the infrastructure.

According to Statista, AWS’s market share is currently at 24.3%, over 8% ahead of IBM Cloud’s 16%.

Google Cloud’s market share at 7.4% is barely larger than Microsoft Azure’s 7%. But on a global scale, these four US-based giants are clear market leaders.

Together, they represent more than 54% of all IaaS and PaaS revenues worldwide.

After the leading US providers, you have China’s Alibaba at 4.4% and Japan’s NTT Data at 2.8% total share of global revenue.

It’s still very much a US-led industry globally, both for actual usage and ongoing innovation.

AWS is the leading IaaS and PaaS provider and continues to grow rapidly into 2021.

What Services Does PaaS Include?

Although the most common use case of PaaS is web app deployment, many other cloud services also fall under it.

Let’s take a closer look.

Database as a Service (DBaaS)

A cloud-hosted database that you manually install on a virtual machine is only an implementation of IaaS.

To be considered a PaaS offering, it needs to be an integrated solution that offers storage, computing power, and relational database capabilities.

An example of this is the Azure SQL Database service, which offers a fully managed database with automated updates, scalability, smart threat protection, and AI-powered search.

Cloud service products in this category are also called DBaaS, a subcategory of PaaS.

Internet of Things (IoT) Platforms

More items are powered by computers and connected to the internet than ever before. The new HTTP/3 standard will only accelerate that further. Connected devices now include lights, thermostats, ovens, washing machines, locks, and even truck engines.

The bare bones of connectivity to the internet could be considered IaaS, but complex APIs for controlling and sharing data across devices and apps fall under PaaS.

Mobile Services (APIs)

Companies are no longer settling for email when sending notifications and marketing campaigns to their customers.

They also use automated SMS messages at scale.

With SMS APIs, companies can build automated messages into their applications.

For example, they can text customers to:

• Remind them of scheduled calls or meetings.
• Promote a new related product or service.
• Ask for feedback on a recent customer service encounter.
• Recruit customers to join a case study or survey.

These services are sometimes categorized separately as Communications Platform as a Service (CPaaS), a PaaS subcategory.

Push Notification APIs

Like SMS text messages, except for browser and mobile push notifications, these APIs power push notifications.

You can use them to:

• Remind customers to install a new update for your app.
• Win back inactive users who haven’t accessed your app in weeks.
• Advertise a relevant limited-time promotion.

Machine Learning

If you genuinely want to take advantage of your data, it’s not enough to just store it in the cloud. The data is still just sitting around, only in a new location.

You need to set up algorithms to sift through your data and find meaningful insights and actionable steps.

With cloud-based machine learning platforms, you can easily create models (from templates), apply them to your databases, and scale your computing power as needed.

For example, IBM Watson Studio lets you automate AI lifecycle management, deploy and run models with a single click, and more.

It’s a great PaaS environment for making use of big data.

AI-powered search and suggestions are also part of the PaaS development tools that the big four offer.

Hadoop, Spark, & Other Data Processing Frameworks

Apache Hadoop is an open source software framework that makes it possible to process big data sets across distributed clusters of virtual machines.

Instead of setting up the environment from scratch, you can use Hadoop as a service from any leading PaaS vendor.

• • Google offers Hadoop as an integrated part of its Dataproc big data processing service.
  • Microsoft offers Hadoop as part of its HDInsight data processing service.
  • IBM offers Hadoop as part of BigInsights.
  • AWS offers Hadoop and Spark as part of EMR.

These leading companies also offer custom data lake and data processing services beyond Hadoop.

Most Popular PaaS Services

Many of the most popular cloud solutions are PaaS services. Just look at these results from a 2020 survey on public cloud services.

Cloud-based relational databases are the most popular, with 67% of companies already using them, 17% experimenting, and 10% planning to use them.

In third place, you have data warehousing. 53% of companies currently use this as a solution for handling and analyzing big data.

Google Cloud BigQuery is an example of this type of PaaS product.

The 4 Leading PaaS Providers: What Services Do They Offer?

As we’ve already covered, within public cloud services, there are four clear market leaders.

But how do they stack up against each other in PaaS service offerings?

Below, we’ll take a closer look at every notable cloud service provider and what they bring to the table.

AWS

AWS is the original cloud computing provider, having launched the revolution with its primary EC2 product in 2006.

The head start cemented them as the clear market leader, and it’s still the largest cloud services company in the world.

But for PaaS specifically, what does it bring to the table?

A quick look at Amazon’s services overview will tell you everything you need to know.

The majority of the highlighted use cases actually represent a PaaS product. Let’s break down exactly what AWS offers in terms of PaaS products.

The days when AWS only offered computing power and virtual machines for rent are long gone.

It now has custom products for every major PaaS service and use case, from app deployment and big data to DBaaS and machine learning.

Always an early mover, Amazon launched a native notification service, Simple Notification Service (SNS), in 2010. That’s the same year Twilio was founded.

If you’re looking for a versatile PaaS provider, you can’t go wrong with the industry’s most experienced veteran.

IBM Cloud

What about IBM Cloud? An early innovator in computing, IBM has put a lot of money and effort into developing its cloud services suite.

IBM first launched its PaaS services as IBM Bluemix in 2014.

In 2017, IBM dropped the Bluemix brand and grouped its PaaS, IaaS, and private cloud offerings under the IBM Cloud umbrella.

With a wide range of enterprise clients, IBM Cloud has quickly grown to become one of the leading PaaS providers since its launch in 2011.

And that shows in its range of services:

But how does IBM stack up in the PaaS department?

IBM Cloud covers essential PaaS use cases like app deployment, big data processing, and data warehousing.

But for notifications, IBM does not offer a complete solution yet. You have to rely on third-party providers like Twilio to introduce SMS into your application workflow.

And IBM has long been on the frontier of machine learning and AI. IBM Deep Blue became the first AI to defeat a world chess champion back in 1997.

The IBM Watson Studio makes it easy to put the Watson AI to work, helping you make sense of your data.

Google Cloud

Google isn’t just a search engine. It’s also one of the leading SaaS companies, with Google Docs, Drive, Gmail, and the entire Google Workspace.

Google also lets you rent the infrastructure and platforms that make it possible to handle billions of visitors every month.

Launched in 2008, Google Cloud was the second major player to enter the market. Its extensive list of products shows why it’s still one of the market leaders.

And for PasS-specific products, things are no different.

For app deployment and development, Google offers a wide range of tools, including the Google App Engine and Kubernetes Engine.

Google also offers many big data processing options through its Dataproc service. You can use Hadoop, Spark, or other frameworks to set up clusters and start processing terabytes of data quickly.

For regular databases and warehousing, Google also offers several options, like BigTable, Google Cloud SQL, and more.

With Firebase Cloud Messaging, you can schedule and send push notifications and in-app messages, but not texts.

Google AI platform offers a user-friendly interface to create, manage, and deploy new machine learning models as quickly as possible.

It’s a complete platform suitable for all your PaaS needs.

Microsoft Azure

Microsoft isn’t just responsible for the operating systems on most desktop and laptop computers around the world.

It also has one of the largest public cloud services collections, including Office 365, Microsoft Teams (SaaS), and Azure (IaaS & PaaS).

The Azure cloud platform includes a range of services from AI and machine learning to analytics, development tools, data processing, and more.

And at the platform level, Microsoft’s got you covered as well.

The Azure App Service makes app deployment easy with built-`in patching, security, scaling, and a host of integrations.

You can store and process data with a range of SQL-based solutions like a data warehouse or a smaller-scale database. You can then use Azure Machine Learning to create and deploy models to make sense of your data.

Azure also supports building automated push and SMS notifications into your app experience.

Summary

PaaS doesn’t just offer you the computing power and infrastructure you need to collect, share, and implement data better across your organization. It also includes the environment needed to take advantage of the data in real-time and implement internal or external applications.

If you’re looking to take advantage of PaaS’s power for your website, without the hassle of finding the best combination of services to create your environment, you can use our free migration service. Our accessible hosting plans rely on cutting edge cloud infrastructure from Google Cloud, without the headache of setting it all up.

If you have any questions about or experiences with PaaS products and services, please drop a line in the comment section!