Redb: Real-Time Enhancement of Docker Containers Via Bank Partitioning in Multicore Systems

Abstract

Docker is a lightweight virtualization technology adopted by numerous companies to develop, deploy and manage applications. However, All Docker containers running on the same system share system resources, leading to performance interference among them. This interference adversely affects the predictability of program response times within the containers. Thus, shared resources isolation between containers has become an important issue. Efforts have been made to improve isolation across various dimensions, such as file systems, I/O, etc. However, consideration of the effect of dynamic random access memory (DRAM) bank interference on the multicore architecture remains open due to the sharing of the DRAM banks among cores. We propose a novel approach, Real-time Enhancement of Docker Containers via Bank Partitioning on Multicore Systems (REDB), as an effective solution to achieve DRAM isolation among containers, thereby enhancing the real-time performance of Docker containers. The key strategy is to partition DRAM banks and allocate specific banks to particular containers. Our evaluations indicate that the REDB approach substantially mitigates DRAM bank interference and significantly improves real-time performance, reducing the maximum latency by 95.3%.

Keywords: bank partitioning, Docker, multicore architecture, isolation, real-time