Cloud Computing with ProfitBricks – Genuine IaaS
Cloud computing describes a concept whereby computing power is purchased as a service via Internet ("Cloud") that gives the user access to virtual resources such as processors, server, RAM, network components or hard disks. At ProfitBricks, these resources are dedicated, which means that they are only available to one user and are physically located in a German data center.
Cloud computing as a more comprehensive concept compared with virtualisation and traditional outsourcingCloud computing is based on the virtualisation of server hardware. This makes it possible to launch multiple virtual servers on a single server, which can be operated independently of one another. This allows for much more efficient use of the available hardware. The concept of cloud computing goes far beyond the pure virtualisation of computing instances in several respects because it enables a general overview of the IT infrastructure and, in particular, allows for demand-driven scaling of computing power.
What’s more, it is assumed in the case of cloud computing that the user of cloud computing services is not also simultaneously monitoring the functioning of the physical infrastructure. Against this background, the concept can also be described as the outsourcing of an infinitely scalable virtual data centre environment.
Learn more about the topic of cloud computing in our cloud lexicon.
Types of Cloud Computing
The most important distinction is between the public cloud and the private cloud. The public cloud is where a provider makes a service available to many customers simultaneously. A private cloud means that a service is provided by dedicated servers that are not shared with others. The combination of both types is referred to as a hybrid cloud. The company that provides a hybrid or public cloud service is called a cloud provider.
The security aspect is regularly cited as an argument in favour of the private cloud. The use of a closed cloud computing environment in the private cloud allows security-relevant data to remain in the company and to be stored in enclosed areas. This prompts many companies to run the majority of their processes in a self-operated private cloud. In doing so, they often ignore the fact that the main cause of loss of data is not physical access to the server, but unauthorized access to data via data connections. The best protection against this risk is not steel and concrete, but an up-to-date security concept and continuous monitoring of data traffic and data security. Private cloud users often fall into a false sense of security and neglect basic safety measures.
The particular advantage of using a public cloud relates to the dynamic scalability of resources. This means that the resources provided can be adapted dynamically to suit current requirements. This eliminates the possibility of insufficient or excess availability of resources, and of costly capacities lying idle.
Many cloud computing services can also be divided into three categories: Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS). These cloud computing services are also referred to as a cloud computing stack. With the right choice of the cloud computing service, companies get the right level of control.
Advantages and possible uses for Cloud Computing
The key benefit of cloud computing is that it provides the user at all times with precisely the required computing and storage capacity or software as a type of self service (Cloud Self-Service). In theory, there should be no bottlenecks in the supply of computing power and storage capacity and, in turn, these resources should not be left “lying idle”. This offers the advantage that the capacity procured in this way using a rental model can actually be billed based on consumption. As a result, IT no longer forms part of a company’s business assets that requires high levels of initial and replacement investment and which must be written off, but can instead be deducted from a company’s ongoing operating costs. High and risky investments in server hardware are no longer required because server performance is purchased, like electricity from the wall outlet, based entirely on demand.
The following use cases are possible for cloud computing:
- Web applications
- Save, back up, and restore data
- Websites and blogs
- Audio and video content streaming
- Big Data analysis
Cloud Computing at ProfitBricks
Here at ProfitBricks, we concentrate as part of our cloud computing services on Infrastructure as a Service (IaaS). This means that we provide customers with computing capacity purely as an infrastructural service. We operate data centres in Germany on which customers can base virtual data centres.
In addition to the usual virtualisation of physical servers, we have developed control software that also enables the virtualisation of the network components of a data centre. ProfitBricks integrates this software at a low level in the network model (level 2 – level 3) so as to automate with the help of software control all communication taking place through direct network connections between the individual components. This technical concept is referred to as Software Defined Networking (SDN). The virtual data centre created in this way is a Software Defined Data Centre (SDDC).
This opens up an unprecedented degree of flexibility and compatibility with structures in data centres with “real” hardware. Therefore, we describe our solution as real cloud computing for the digital age, in contrast to the offerings of other providers that still tend to be oriented more towards conventional dedicated hosting concepts, where the term “Cloud” is simply applied to what are in effect traditional hosting methods.
In practice, we break down virtual instances into their smallest possible component parts (cores, RAM and storage) before reassembling to meet current demand. This approach produces almost endless configuration possibilities that comprise cores, RAM and storage, beginning with 1 core, 0.25 GB RAM and 10 GB storage and extending all the way to a current maximum size of 62 cores, 240 GB RAM and 16 TB per storage device (up to eight per server) per virtual instance. This ensures that the customers can configure their virtual servers precisely to ensure optimum and cost-efficient processing of their workloads. This also means that scaling is not only implemented by the addition of new instances (horizontal scaling), but also by the vertical scaling of individual servers during live operations (Live Vertical Scaling – LVS). This means that the individual parameters of a virtual server can be raised or lowered independently of one another – during the runtime of the systems i.e. without having to shut down and restart the virtual server. A real benefit in terms of reliability. In addition, technologies such as IP failover ensure even more clearly raised availability.