/ Glossary / By

Cloud hosting explained

Cloud hosting is a sub-term for cloud computing and describes the provision of storage space from a cloud provider’s data center. Cloud computing includes Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS) offerings. Cloud hosting services are the further development of so-called legacy hosting services and differ from these in key respects.

Different types of hosting

There are different types of hosting. Below we compare the different services with each other to highlight the individual advantages and disadvantages.

Shared hosting and cloud hosting

With shared hosting, the websites are located on local computers, whereby the computers are networked. Although cloud hosting is based on local computers from which the resources are made available, the websites are designed for a virtualized environment. The main disadvantage of physical servers is the lack of scalability of server capacities.

Public and private cloud hosting

With public cloud hosting, several customers can use resources from the same cloud environment and scale them as required. Private cloud hosting is also known as dedicated private cloud hosting. The computer capacities are exclusively available to a single group of users, usually companies. As the private cloud is not accessible to other users outside the company, a high level of security and control is guaranteed. Access to the cloud can be achieved either via a protected intranet or via a VPN. A private cloud is therefore a cloud that is only accessible to a specific company and its internally authorized users.

Cloud hosting or dedicated hosting?

Cloud hosting is the logical development of the dedicated hosting and virtual server hosting sectors. Thanks to modern virtualization, cloud solutions have now developed into a real added value compared to on-premise architectures.

Advantages of cloud hosting

Modern cloud hosting offers the following advantages:

  • Any number of complex network configurations thanks to Software Defined Networking (SDN)
  • Computing power can be adapted to any requirements in real time without the need to restart a cloud server.
  • There is no contractual commitment and therefore no risk for the user. The user can decide at any time to purchase more or less (or no) computing power.
  • Computing power is billed by the minute according to consumption, i.e. exclusively pay-as-you-go.
  • By using high-quality enterprise hardware and intelligent distribution systems, the cloud hosting provider guarantees that the user’s infrastructure is highly available.

How does a dedicated hosting provider work?

A dedicated hosting provider usually has a large number of dedicated servers preconfigured (CPU, RAM and hard disk) and installed in its data center. The provider then rents out these servers on the basis of a term contract. Typical contract terms are between one and 24 months.
After an order is placed (usually via a type of store system), the provider automatically selects a suitable free dedicated server and installs it with an operating system. Once the installation is complete, the user is informed by e-mail. From this point on, the user can access the server via the Internet and maintain it remotely.

Disadvantages of dedicated servers

However, the provider’s savings in the choice of hardware come at the expense of reliability. For example, consumer hardware often lacks redundant power supplies, ECC memory and a cooling system that is designed for 24/7 operation in a data center. When it comes to hard disks, providers often opt for so-called “green models” that only rotate at low speed. This allows the provider to save considerable amounts of electricity, but the user suffers from significantly slower access speeds to their data.
There are a number of disadvantages, particularly when the dedicated server is compared with modern cloud services:

  • If a piece of hardware fails, there is a long downtime. This is because the provider must first replace the hardware in order to rectify the defect.
  • The computing capacities are limited to one physical server. An application cannot request additional CPU or RAM capacities.
  • The contract terms for most dedicated servers are often long and inflexible.
  • Private networks between several dedicated servers are not possible.
  • The computing capacity is static and it is difficult to determine the actual demand.
  • Due to the non-scalable pre-programmed server for high traffic, the server capacities are largely underutilized during normal use, which means that more resources are paid for than are used.
  • If consumer hardware is used instead of enterprise hardware, it is not usually suitable for 24/7 operation and its functionality and durability will be impaired.

Conclusion

The key advantage of a cloud hosting provider is that it provides computing and storage capacities in line with demand and these capacities can be adapted to other performance requirements at any time. Using intelligent distribution algorithms within a cloud system, the provider also ensures that the workloads are distributed evenly across the available servers. This greatly reduces the probability of a failure.

Matching products