What are the disadvantages of IBM Bluemix

Technical advantages and disadvantages of the cloud

Clouds have several advantages over own data centers. The resources are almost infinite. Hardware resources such as computing power (CPUs, GPUs, FPGAs), RAM and storage can be increased within a short period of time - depending on the product and service, it can take seconds to hours. The resources can also be scaled almost in real time. There are often cases in which a particularly high amount of power is required at certain times or seasons. Think, for example, of batch runs at certain times of the day or of the particularly high number of customers at Christmas time. In the cloud, the required resources can be scaled up or down as required. With some components this even works almost in real time. That means: While a batch data run is being carried out, you can increase the performance class of the required component in the cloud immediately before particularly compute-intensive operations and reduce the performance class after the computation-intensive operations have been completed. This also allows you to optimize costs by trying to allocate resources according to current needs.

  1. V-LINE EUROPE GmbH, Sehnde
  2. University of Applied Sciences Bonn-Rhein-Sieg, Sankt Augustin

In the cloud, hardware components are available for special projects that would be difficult or impossible to get in your own data center because they are not part of the company's in-house standard. Think, for example, of a new machine learning project that requires TESLA acceleration cards from Nvidia or FPGAs from Xilinx, but which are not available as standard components in your own company. Exotic operating systems or tools can be tested in isolated networks in the cloud without endangering your own company network due to potential incompatibilities or violating company-internal security guidelines. You can keep data redundant in the cloud worldwide and thus enable short access times for customers and partners on site. For many companies, this is more cost-effective than setting up their own data centers in distant countries. The company itself can focus on its core competencies. A trading company could ask the question: Do we absolutely need our own large data center or should we concentrate on our actual business, trading, and buy the primary computer capacities in the cloud? That doesn't mean that you shouldn't have your own, smaller data center in your own building. You can combine both, the private data center and the resources in the cloud.

Disadvantages of the cloud - internet connection and privacy

Despite the advantages, the cloud is not suitable for everything, its use also has certain disadvantages. The cloud resources are almost infinite, but there is often a bottleneck: the often limited internet connections between companies and cloud providers as well as the response times that result from the physical distance. In cases where the limitation is not the computing power or the memory, but the internet bandwidth and response times, it is worthwhile from a performance point of view to keep certain data locally in your own data center so that it can be accessed more quickly within the LAN.

When uploading data to the cloud and / or downloading data from the cloud, there are traffic costs that can vary depending on the cloud provider. These costs must be taken into account in the calculation. The data protection aspects must not be forgotten either. First of all, these have nothing to do with data security. A system in the cloud, if well configured, can even be more secure than some systems in your own company network. However, the data is stored in the cloud by a third-party service provider and this must be checked under data protection law. Another aspect is the geographic location of the cloud data centers. With some cloud providers, you can restrict these according to country or region and thus ensure, for example, that the data is stored within the EU, which can be significant in some cases from a data protection perspective.

  1. OpenShift installation & administration
    14.-16. June 2021, online
  2. Terraform with AWS
    14./15. September 2021, online
Further IT training

Regarding the last point: A general distinction is made between public, hybrid and private cloud. The public cloud is offered publicly on the Internet by a cloud provider. The data is stored in various data centers, mostly distributed over several countries, which hold data from many customers at the same time, but can separate them from one another using authorization concepts. Examples of such cloud offers are Google Docs or Microsoft Office 365. The opposite of the public cloud is the private cloud, in which the data and applications are kept in the customer's own local data centers or on individually assigned servers of a cloud provider. In some cases, this can be interesting from a data protection perspective.

Some cloud providers also offer so-called hybrid cloud solutions. Certain components or data can be stored privately in your own data center instead of in the public cloud. This not only avoids data protection problems, but also optimizes performance. The above advantages and disadvantages generally apply to all clouds. Regarding one or the other aspect, however, there are definitely differences between the cloud providers.

  2. 1
  3. 2
  4. 3
  5. 4
  6. 5