Hi Leo,O
Glad to hear from you, I am an active member of KubeEdge as well as member of ZJU-SEL. First, I would like to show you our understanding of OpenEuler and edge-IOT, so that we could engage in deeper communication.
The story begins sevaral months ago, we meet a need to manage a large cluster of android devices, so we investigated possible devices and operating systems running on them in edge scenarios :
- Relatively high performence processor, Cortex-A, OS(GUN/Linux distribution, Android, Android Things)
- example, RPi
- application, home gateway box, TV box, UAV
- Low Power MCU, Cortex-M, OS(LiteOS, AliOS Things, TencentOS-Tiny),
- example MSP430
- application, shared bikes, intelligent street lamp
To my view, OpenEuler claims it is based on centOS, so I think it fits the first case. For now, Android or Android Things is not able to run docker or containers because its kernel has been cropped, and GUN/Linux distribution is too heavy to run at edge. At the same time, our competitor--k3s has released its assorted OS called K3OS. So to sum up, oepnEuler is beneficial for KubeEdge or EdgeComeputing:
- an OS support run containers to competed with Android, to broaden the boundaries of kubeedge deploying.
- an optimized OS is definitly helpful for KubeEdge users,
improve our competitiveness.
And the last, what KubeEdge cloud do in this SIG?
- verify intigration between OpenEuler and KubeEdge, such as iSula?
- join the meetings of OpenEuler?
- introduce OpenEuler at KubeEdge meeting?
Thx