一句话：提出 Pond，首个在满足云性能目标的同时显著降低 DRAM 成本的 CXL 内存池化系统。
公有云需在严格性能要求下降低硬件成本，而主存是关键。内存池化可提升 DRAM 利用率以降低成本，但在云性能目标下颇具挑战。本文提出 Pond，首个既满足云性能目标又显著降低 DRAM 成本的内存池化系统。Pond 基于 Compute Express Link（CXL）提供对池内存的 load/store 访问，并利用两个洞察：云端生产 trace 分析表明，跨 8–16 个 socket 池化即可获得大部分收益，从而支持低延迟小池设计；机器学习模型可准确预测应为虚拟机分配的本地与池内存，以逼近同 NUMA 节点性能。基于 158 个工作负载的评估显示，Pond 降低 DRAM 成本 7%，性能与同 NUMA 节点 VM 分配仅差 1–5%。
Public cloud providers seek to meet stringent performance requirements and low hardware cost. A key driver of performance and cost is main memory. Memory pooling promises to improve DRAM utilization and thereby reduce costs. However, pooling is challenging under cloud performance requirements. This paper proposes Pond, the first memory pooling system that both meets cloud performance goals and significantly reduces DRAM cost. Pond builds on the Compute Express Link (CXL) standard for load/store access to pool memory and two key insights. First, our analysis of cloud production traces shows that pooling across 8-16 sockets is enough to achieve most of the benefits. This enables a small-pool design with low access latency. Second, it is possible to create machine learning models that can accurately predict how much local and pool memory to allocate to a virtual machine (VM) to resemble same-NUMA-node memory performance. Our evaluation with 158 workloads shows that Pond reduces DRAM costs by 7% with performance within 1-5% of same-NUMA-node VM allocations.
@inproceedings{paperbot10051,
title = {Pond: CXL-Based Memory Pooling Systems for Cloud Platforms},
author = {Huaicheng Li and Daniel S. Berger and Lisa Hsu and Daniel Ernst and Pantea Zardoshti and Stanko Novakovic and Monish Shah and Samir Rajadnya and Scott Lee and Ishwar Agarwal and Mark D. Hill and Marcus Fontoura and Ricardo Bianchini},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3575693.3578835}
}