Abstract: Toward Operating System Assisted Hierarchical Memory Management for Heterogeneous Architectures

Heterogeneous architectures, where a multicore processor is accompanied with a large number of simpler, but more power-efficient CPU cores optimized for parallel workloads, are receiving a lot of attention these days. Currently, these co-processors come with a limited on-board memory, which requires partitioning computational problems manually into pieces that can fit into the device´s RAM as well as overlapping computation and communication. In this poster we propose application transparent, operating system (OS) assisted hierarchical memory management system, where the OS orchestrates data movement between the host and the device and updates the process virtual memory address space accordingly. We identify the main scalability issues of frequent address space changes, such as the increasing price of TLB invalidations with the growing number of CPU cores and propose partially separated page tables to overcome the problem. With partially separated page tables each core maintains its own set of mappings of the computation area, enabling the OS to perform address space updates in a scalable manner, and involve a particular CPU core in TLB invalidation only if it is absolutely necessary. Furthermore, we propose dedicated data movement cores in order to efficiently overlap computation and communication. We provide experimental results on stencil computation, a common HPC kernel.