Efficient Memory Disaggregation with Infiniswap

Authors: 

Juncheng Gu, Youngmoon Lee, Yiwen Zhang, Mosharaf Chowdhury, and Kang Shin, University of Michigan

Abstract: 

Memory-intensive applications suffer large performance loss when their working sets do not fully fit in memory. Yet, they cannot leverage otherwise unused remote memory when paging out to disks even in the presence of large imbalance in memory utilizations across a cluster. Existing proposals for memory disaggregation call for new architectures, new hardware designs, and/or new programming models, making them infeasible.

This paper describes the design and implementation of INFINISWAP, a remote memory paging system designed specifically for an RDMA network. INFINISWAP opportunistically harvests and transparently exposes unused memory to unmodified applications by dividing the swap space of each machine into many slabs and distributing them across many machines’ remote memory. Because one-sided RDMA operations bypass remote CPUs, INFINISWAP leverages the power of many choices to perform decentralized slab placements and evictions.

We have implemented and deployed INFINISWAP on an RDMA cluster without any modifications to user applications or the OS and evaluated its effectiveness using multiple workloads running on unmodified VoltDB, Memcached, PowerGraph, GraphX, and Apache Spark. Using INFINISWAP, throughputs of these applications improve between 4x (0:94x) to 15:4x (7:8x) over disk (Mellanox nbdX), and median and tail latencies between 5:4x (2x) and 6x (2:3x). INFINISWAP achieves these with negligible remote CPU usage, whereas nbdX becomes CPU-bound. INFINISWAP increases the overall memory utilization of a cluster and works well at scale.

NSDI '17 Open Access Videos Sponsored by
King Abdullah University of Science and Technology (KAUST)

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

Presentation Video

Download Video

Presentation Audio

BibTeX
@inproceedings {201565,
author = {Juncheng Gu and Youngmoon Lee and Yiwen Zhang and Mosharaf Chowdhury and Kang G. Shin},
title = {Efficient Memory Disaggregation with Infiniswap},
booktitle = {14th {USENIX} Symposium on Networked Systems Design and Implementation ({NSDI} 17)},
year = {2017},
isbn = {978-1-931971-37-9},
address = {Boston, MA},
pages = {649--667},
url = {https://www.usenix.org/conference/nsdi17/technical-sessions/presentation/gu},
publisher = {{USENIX} Association},
}