Rethinking Memory Profiling and Migration for Multi-Tiered Large Memory Systems

TL;DR

This paper introduces MTM, a novel memory management system for multi-tiered large memory architectures that improves profiling, migration, and performance, especially for big-data applications with large working sets.

Contribution

MTM presents a unified, application-transparent approach to memory profiling and migration tailored for complex multi-tiered large memory systems, incorporating huge page awareness.

Findings

MTM outperforms seven state-of-the-art solutions by up to 42%.

MTM achieves an average performance improvement of 17%.

MTM effectively manages large working sets in big-data applications.

Abstract

Multi-tiered large memory systems call for rethinking of memory profiling and migration because of the unique problems unseen in the traditional memory systems with smaller capacity and fewer tiers. We develop MTM, an application-transparent page management system based on three principles: (1) connecting the control of profiling overhead with the profiling mechanism for high-quality profiling; (2) building a universal page migration policy on the complex multi-tiered memory for high performance; and (3) introducing huge page awareness. We evaluate MTM using common big-data applications with realistic working sets (hundreds of GB to 1 TB). MTM outperforms seven state-of-the-art solutions by up to 42% (17% on average)