Memory tiering performance improved with machine learning predictions
MANTA: Machine Learning Augmented Tiering Advisor
Distributed, Parallel, and Cluster ComputingOperating Systems
Summary
Memory in datacenters can be split into fast and slow parts to balance speed and size. The challenge is keeping the most useful data in the fast memory as workloads change rapidly. The authors studied current methods and developed a way to predict which data will be useful soon. Their system, MANTA, uses these predictions to manage memory better and showed speed improvements in tests with different workloads and memory types.
What this means in practice
- •For datacenter operators: Implement MANTA to improve memory management efficiency and reduce slowdowns caused by changing workload patterns.
- •For systems software developers: Integrate machine learning models into memory tiering policies to dynamically optimize page placement with minimal overhead.
Authors
Johannes Freischuetz, Kiet Pham, Sujay Yadalam, Konstantinos Kanellis, Michael Swift, Shivaram Venkataraman
Abstract
Memory tiering has been used to expand memory capacity, particularly in datacenters, by combining fast DRAM with slower tiers, including CXL-attached memory. Its effectiveness depends on keeping useful pages in the fast tier, but existing heuristic policies can lag behind changing hot sets in phased or bursty workloads. To explore these limitations, we introduce ChOMP, a scalable offline optimizer that minimizes placement and bandwidth-sensitive migration costs. We then develop a trace-driven simulator that uses this reference to identify performance opportunities for online policies. Motivated by these results, MANTA predicts future page usefulness from runtime access features and integrates a lightweight learned model into ARMS. Across eight workloads on emulated CXL, MANTA achieves geometric-mean speedups over ARMS of 1.12$\times$ and 1.08$\times$ at 4~GB of fast memory on Linux 6.2 and 6.18, respectively; across six Optane workloads, it achieves 1.69$\times$. On individual workloads, MANTA is up to 1.25$\times$ faster than ARMS with emulated CXL on Linux 6.2, 1.21$\times$ on Linux 6.18, and 5.6$\times$ with Optane.