MLB: A Memory-aware Load Balancing Method for Mitigating Memory Contention

Abstract

Most of the current CPUs have not single cores, but multicores integrated in the Symmetric MultiProcessing (SMP) architecture, which share the resources such as Last Level Cache (LLC) and Integrated Memory Controller (IMC). On the SMP platforms, the contention for the resources may lead to huge performance degradation. To mitigate the contention, various methods were developed

most of these methods focus on finding which tasks share the same resource assuming that a task is the sole owner of a CPU core. However, task arrival patterns and the demand for resources in the current server environment are highly dynamic

hence, tasks, the number of which is larger than that of CPU cores, can be executed simultaneously on a CPU. In order to mitigate contention for memory subsystems in such multitasking cases, dealing with the dynamicity of resource demand, we have devised a Memory-aware Load Balancing (MLB) method. MLB dynamically recogrnizes contention by using simple contention models and performs inter-core task migration to mitigate the contention. We have evaluated MLB on an Intel i7-2600 (desktop-level CPU) and a Xeon E5-2690 (server-level CPU), and found that our approach can be effectively taken in an adaptive manner, leading to noticeable performance improvements of memory intensive tasks (about 15% in best case) on the different CPU platforms. Also, MLB can achieve performance improvements in CPU-GPU communication in discrete GPU systems.