This paper focuses on a distributed cooperative guidance model-free control method for a cluster of under-actuated disk-type autonomous underwater gliders (AUGs) in the presence of unknown kinetic model parameters and ocean disturbances. Firstly, a distributed cooperative motion generator is designed to generate reference path points, and then a cooperative control method is proposed based on the update path parameters. Secondly, a three-dimensional (3D) guidance law is constructed by employing closed 3D vector fields. Finally, data-driven filtered adaptive extended state observers (DFAO) are proposed to deal with the unknown input gains, internal uncertainties and external disturbances of the disk-type AUGs, and an adaptive kinetic control law is designed by using the knowledge learned from the observers. Simulation results demonstrate the effectiveness of the proposed 3D distributed cooperative guidance model-free control method for disk-type AUGs subject to fully unknown kinetics.
Containerization is becoming increasingly popular, but unfortunately, containers often fail to deliver the anticipated performance with the allocated resources. In this paper, we first demonstrate the performance variance and degradation are significant (by up to 5x) in a multi-tenant environment where containers are co-located. We then investigate the root cause of such performance degradation. Contrary to the common belief that such degradation is caused by resource contention and interference, we find that there is a gap between the amount of CPU a container reserves and actually gets. The root cause lies in the design choices of today's Linux scheduling mechanism, which we call Forced Runqueue Sharing and Phantom CPU Time. In fact, there are fundamental conflicts between the need to reserve CPU resources and Completely Fair Scheduler's work-conserving nature, and this contradiction prevents a container from fully utilizing its requested CPU resources. As a proof-of-concept, we implement a new resource configuration mechanism atop the widely used Kubernetes and Linux to demonstrate its potential benefits and shed light on future scheduler redesign. Our proof-of-concept, compared to the existing scheduler, improves the performance of both batch and interactive containerized apps by up to 5.6x and 13.7x.