When you search for information in regard to uptime, you will quickly come across terms such as IaaS, PaaS and SaaS. They are the respective abbreviations for Infrastructure-as-a-Service, Platform-as-a-Service and Software-as-a-Service. Thanks to Nucleus, a while ago the concept of Uptime-as-a-Service was added to this list. But where do the differences lie?
All four have “as-a-Service” in common. After all, the main aspect is that you are offered something “as a service”. But what do you receive exactly? In order to explain this, first you must look at the different layers in which the different terms can be found. We preferably like to present these layers as an upside-down pyramid
You might consider it to be an unstable model because it has not been based on a stable foundation, but you should rather see it as an unfolded fan, in which a narrow yet firm grip will provide considerable freedom of movement at the wide top.
The sturdy bottom is formed by Infrastructure-as-a-Service (IaaS). This simply replaces your own infrastructure. Instead of buying servers, storage firewalls, network equipment, etc. yourself, from now on you rent this “as a service”.
As a result, IaaS provides the most freedom. You still determine what happens with your infrastructure however: you can (usually) select which operating system and which software is installed on the infrastructure. You simply rent infrastructure in the cloud and manage your virtual servers with your own system administrators or your IT supplier.
The advantages lie in the fact that you no longer have any worries in regard to investments in hardware, maintenance costs, high availability and back-ups. Essentially, there are very few disadvantages connected with this manner of operation, provided the infrastructure is well maintained.
With IaaS there are various types of subscriptions, which can be divided into two main groups: Elastic and dedicated clouds.
Elastic clouds are elastic and scalability is their most important characteristic. You rent processing power, storage and bandwidth and you only pay for what you actually use. The consumption is settled in CPU cycles, read/write movements, etc.
The main advantage of course is that applications will virtually instantaneously and automatically be allocated more resources when they require them. The main disadvantage is that you will have no knowledge beforehand of the costs involved. An additional disadvantage is that firm guarantees for this elasticity cannot always be provided.
With dedicated clouds, your capacity can be determined beforehand. Your servers will be reserved (dedicated) for you in the cloud. As a result, you can be sure that this capacity will always be present for you alone. Simultaneously, you maintain the advantage that you control the costs, but with a single push of a button can still expand the capacity, yet without encountering financial surprises.
IaaS is usually also the supplier for the layers that will be constructed above this, such as PaaS and SaaS.
The next layer is Platform-as-a-Service (PaaS). An additional layer is constructed above the infrastructure, on which application developers can work in order to program their software in the cloud and then offer it as a SaaS solution. This layer is almost exclusively targeted at programmers and is a true intermediate layer.
The main disadvantage of PaaS services is the vendor lock-in. When you develop an application on a specific PaaS platform, then you are often bound to this, because each platform has developed its own standard. As a result, applications rarely or possibly even never can be moved from one platform to the other.
For software developers, this layer has the advantage that they no longer have to worry about system management, scalability, etc. and that they can be sure that their application will then work for everyone in the same manner in a SaaS model.
Most PaaS standards provide developers with tools and guidelines that must enable them to develop applications, which can be enormously scalable. Furthermore, the communication between the different applications on the same platform is performed much more easily.
Because the PaaS platform itself ensures the scalability, the application will always operate equally well, irrespective if only ten or perhaps ten thousand users are using the application.
The largest layer is Software-as-a-Service. These are applications that are not installed on your computer, but instead run in the cloud. The main advantage is that you work irrespective of the type of device or location.
Usually, with SaaS you will pay a monthly subscription, often based on the number of users. This amount is all inclusive. SaaS providers ensure that their application operates in the cloud (by using IaaS or PaaS – possibly self-maintained). You also benefit from always having the most recent versions of the software at your disposal.
However, the disadvantage is that large SaaS applications have a certain degree of uniformity. The development of software is quite costly and the more users over which these costs can be amortised, the less expensive the applications become. As a result, SaaS will mainly be successful for applications that have a large number of users, such as e-mail, sales collaboration, CRM, ERP, etc.
The advantages lie in the fact that for a fixed amount you will be spared from most worries and you can work everywhere and at any time.
A while ago, Nucleus added an additional term to the list, namely Uptime-as-a-Service (UaaS). We also like to describe this as IaaS Plus.
Because although many companies see the benefit of Infrastructure-as-a-Service, they continue to struggle with the fact that the leased infrastructure must still be maintained by their own system administrators or an IT supplier. With Uptime-as-a-Service, we remove this last hurdle: We provide managed services on top of the leased infrastructure.
At Nucleus, Uptime-as-a-Service means that our experts take care of the management, maintenance, security and online privacy, monitoring and back-ups of your infrastructure. We offer rock solid guarantees for maximal uptime, but make sure we have business continuity and disaster recovery covered as well. We offer our advice before, during and after each project. We’ll also help you with DevOps, allowing developers to build applications and websites that are better aligned with your infrastructure. That way you don’t have to worry about uptime anymore.
Have you got experience with IaaS, PaaS, SaaS or UaaS? Let us know in the comments below.
This blogpost is part of a free eBook that covers all aspects of uptime.