Search
ICNAP is a collaborative center where we come together to address and solve the challenges of digitalization
66
Projects
14
Services
48
Files
7
Topic Fields
70
Scientific Contributors
4
Institutes
To fully leverage the benefits of our resources, we invite you to become a member of our community. Membership provides you with exclusive access to all project results, allowing you to delve deeper into the world of ICNAP.
n the scope of the study „Virtualization for hardware independent deployment of industrial applications” the possibilities and boundaries of virtualization for an implementation of real time applications were investigated. With virtualization the industrial applications can be deployed dynamically on different hardware systems. The advantages of different hardware systems, like computing power, memory and disk space can be used per need.
| Topic Fields | |
| Published | 2021 |
| Involved Institutes | |
| Project Type | ICNAP Community Study |
| Result Type | |
| Responsibles |
Contact us to get in touch! With a membership, you’ll gain full access to all project information and updates.
The most crucial point in virtualization is the latency added on top, by abstracting the software layer. There are different methods to lower the latency of these systems. In this study the configuration of the hypervisor for virtual machines was implemented and tested. By configuring the hypervisor, the assigned hardware resources like the execution orders in the processor can be optimized. With this process called CPU pinning it was able to lower the latency of a virtualized system by ~74%. By mapping these results on the Fraunhofer Edge Cloud a maximum latency of ~723us could be achieved. This value is for sure not close to native real time systems. As most of the native real time systems are used for hard real time applications like motion control it is understandable that the latency needs to be in a very low us range. But there are a lot of applications which are not necessarily running below the 1ms latency range. For this type of applications, the FEC or other cloud systems can be used and integrated into the process.
© Fraunhofer 2025
