Containerization is widely used for isolation in various applications because it is lightweight, scalable, and portable.In modern distributed systems, seamless inter-process communication (IPC) between multi-platform containers is essential for a range of applications and services, including microservices, cloud computing, and Internet of Things (IoT) devices.However, secure and efficient communication between containers on the same host is challenging, especially when different operating systems are involved.This paper introduces Bindox, a lightweight, efficient, and secure IPC mechanism that enables seamless communication across multiple platforms, including Android and Linux.Bindox uses shared memory for data transfer and implements a stable client-server architecture, ensuring high performance and ease of maintenance.Additionally, Bindox provides a robust security mechanism that guarantees confidentiality, integrity, and availability of the communication channel.Experimental results demonstrate that Bindox outperforms existing networking and IPC methods in terms of memory use, latency, and CPU usage, making it a promising solution for efficient and secure communication between multi-platform containers.

GPUs have become ubiquitous in the Cloud-rendering areas due to the outstanding rendering performance. However, many existing Cloud-rendering systems suffer from low GPU utilization caused by the CPU bottleneck. Recent proposals (e.g., API-forwarding and c-GPU) for GPU-usage optimization are promising but fail to address the system-resource redundancy issues (i.e., each instance tends to occupy all the system resources exceeding their requirements), leading to unnecessary CPU consumption and lowering GPU utilization. We conducted an experiment by testing real-world applications on the percentage of unused resources to demonstrate the severity of this issue. Nearly 50% of resources are unused.