Feasibility constraints for edge resources which can be addressed by the PrEstoCloud solution

Edge nodes are resources which can process all the tasks instantly. However, they have relatively feasibility constraints when compared to the infrastructures in cloud-based datacentres. In scenarios of real-time data-intensive Big Data applications orchestrated upon edge computing frameworks, the following constraints are taken into account by the PrEstoCloud solution:

  • Edge devices are low power: One aim of edge computing solutions is to leverage the edge for some computation and take advantages of low-latency. However, there is a constraint that edge devices are characterised by small computing power and they have hardware resource limitations in practice.
  • Edge devices need to be located in the proximity of end-users: In order to benefit from a low-latency response time for the application, edge devices need to be located in the proximity of end-users. This is the main reason that edge computing solutions allow the utilisation of the extreme edge of the network in order to run applications at run-time.
  • Internet connection: Edge devices should be designed for continuous operations in execution environments with preferably a permanent Internet connection. If the edge device has no internet connection for a while, the end-users have no access to the data stored on the cloud. With no reliable Internet connection, possibilities may be limited to collecting data on the edge device, then automatically transporting it to the Internet for further data processing when the edge device gets connected to the Internet.
  • Limited wireless bandwidth: Mobile edge devices usually are connected to the Internet through a low-bandwidth wireless network technology; however the system needs to allow the transmission of real-time streaming data such as video or audio. Minimising the latency through optimal utilisation of bandwidth will be reached with efficient deployment of application fragments.
  • Conflicting desires of end-users: Various users may share an edge node in one region and they may have different desires for example conflicting desires of security and speed. To this end, edge computing solutions may be able to find the best trade-off between different benefits of the technique to process the real-time Big Data and its detriments.
  • Demand for improving data storage capacity: Storage capacity has been considered always as a constraint for edge devices. To overcome this issue, it is possible to store and access the major portion of large data on the cloud through networks. In this way, considerable amount of storage space on edge devices can be saved because it can be sent and processed on the cloud.