On Multi-Level Preemption in Ethernet
Ref: CISTER-TR-180509 Publication Date: 3 to 6, Jul, 2018
On Multi-Level Preemption in EthernetRef: CISTER-TR-180509 Publication Date: 3 to 6, Jul, 2018
Ethernet is increasingly being considered as the solution to high bandwidth requirements in the next generation of timing critical applications that make their way in cars, planes or smart factories to mention a few examples. Until recently, ethernet frames used to be transmitted exclusively in a nonpreemptive manner. That is, once a frame starts transmitting on a switch output port, its transmission cannot be interrupted by any other frame until completion. This constraint may cause time critical frames to be blocked for long periods of time because of the transmission of non-critical frames. The IEEE 802.3br standard addressed this issue by introducing a one-level ethernet frame preemption paradigm. In this approach, frames transmitted through a switch output port are classified as express frames or preemptable frames, depending on their priority levels. Express frames can preempt preemptable frames and two frames belonging to the same class cannot preempt each other. While this partially solves the problem for express frames, all preemptable frames can still suffer blocking irrespective of their priority level. In this work, we investigate the feasibility and advantages of multi-level preemptions in time-sensitive ethernet networks.
30th Euromicro Conference on Real-Time Systems (ECRTS 2018), WiP, pp 16-18.