STYX: Exploiting SmartNIC Capability to Reduce Datacenter Memory Tax

被引:0
|
作者
Ji, Houxiang [1 ]
Mansi, Mark [2 ]
Sun, Yan [1 ]
Yuan, Yifan [3 ]
Huang, Jinghan [1 ]
Kuper, Reese [1 ]
Swift, Michael M. [2 ]
Kim, Nam Sung [1 ]
机构
[1] Univ Illinois, Champaign, IL 61820 USA
[2] Univ Wisconsin Madison, Madison, WI USA
[3] Intel Labs, Champaign, IL USA
来源
PROCEEDINGS OF THE 2023 USENIX ANNUAL TECHNICAL CONFERENCE | 2023年
关键词
D O I
暂无
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Memory optimization kernel features, such as memory deduplication, are designed to improve the overall efficiency of systems like datacenter servers, and they have proven to be effective. However, when invoked, these kernel features notably disrupt the execution of applications, intensively consuming the server CPU's cycles and polluting its caches. To minimize such disruption, we propose STYX, a framework for offloading the intensive operations of these kernel features to SmartNIC (SNIC). STYX first RDMA-copies the server's memory regions, on which these kernel features intend to operate, to an SNIC's memory region, exploiting SNIC's RDMA capability. Subsequently, leveraging SNIC's (underutilized) compute capability, STYX makes the SNIC CPU perform the intensive operations of these kernel features. Lastly, STYX RDMA-copies their results back to a server's memory region, based on which it performs the remaining operations of the kernel features. To demonstrate the efficacy of STYX, we re-implement two memory optimization kernel features in Linux: (1) memory deduplication (ksm) and (2) compressed cache for swap pages (zswap), using the STYX framework. We then show that a system with STYX provides a 55-89% decrease in 99(th)-percentile latency of co-running applications, compared to a system without STYX, while preserving the benefits of these kernel features.
引用
收藏
页码:619 / 633
页数:15
相关论文
共 11 条
  • [1] Making Sense of Using a SmartNIC to Reduce Datacenter Tax from SLO and TCO Perspectives
    Huang, Jinghan
    Lou, Jiaqi
    Sun, Yan
    Wang, Tianchen
    Lee, Eun Kyung
    Kim, Nam Sung
    2023 IEEE INTERNATIONAL SYMPOSIUM ON WORKLOAD CHARACTERIZATION, IISWC, 2023, : 28 - 42
  • [2] Exploiting Data Locality in Memory for ORAM to Reduce Memory Access Overheads
    Kuang, Jinxi
    Shen, Minghua
    Lu, Yutong
    Xiao, Nong
    PROCEEDINGS OF THE 59TH ACM/IEEE DESIGN AUTOMATION CONFERENCE, DAC 2022, 2022, : 703 - 708
  • [3] Exploiting frequency correlation in LTE to reduce HARQ memory
    Torrea-Duran, Rodolfo
    Li, Min
    Desset, Claude
    Pollin, Sofie
    Van der Perre, Liesbet
    2012 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 2012, : 4804 - 4809
  • [4] Exploiting an infrastructure IP to reduce memory diagnosis costs in SoCs
    Bernardi, P
    Grosso, A
    Rebaudengo, M
    Reorda, MS
    ETS 2005:10TH IEEE EUROPEAN TEST SYMPOSIUM, PROCEEDINGS, 2005, : 202 - 207
  • [5] Exploiting the locality of memory references to reduce the address bus energy
    Musoll, E
    Lang, T
    Cortadella, J
    1997 INTERNATIONAL SYMPOSIUM ON LOW POWER ELECTRONICS AND DESIGN, PROCEEDINGS, 1997, : 202 - 207
  • [6] Exploiting Program Cyclic Behavior to Reduce Memory Latency in Embedded Processors
    Atoofian, Ehsan
    Baniasadi, Amirali
    APPLIED COMPUTING 2008, VOLS 1-3, 2008, : 1482 - 1486
  • [7] Exploiting stability to reduce time-space cost for memory tracing
    Gao, XF
    Snavely, A
    COMPUTATIONAL SICENCE - ICCS 2003, PT III, PROCEEDINGS, 2003, 2659 : 966 - 975
  • [8] Exploiting Storage Class Memory to Reduce Energy Consumption in Mobile Multimedia Devices
    Park, Heekwon
    Baek, Seungjae
    Choi, Jongmoo
    Lee, Donghee
    Noh, Sam H.
    2010 DIGEST OF TECHNICAL PAPERS INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS ICCE, 2010,
  • [9] Exploiting k-constraints to reduce memory overhead in continuous queries over data streams
    Babu, S
    Srivastava, U
    Widom, J
    ACM TRANSACTIONS ON DATABASE SYSTEMS, 2004, 29 (03): : 545 - 580
  • [10] Pin or Fuse? Exploiting Scratchpad Memory to Reduce Off-Chip Data Transfer in DNN Accelerators
    Jeong, Hyuk-Jin
    Yeo, JiHwan
    Bahk, Cheongyo
    Park, JongHyun
    PROCEEDINGS OF THE 21ST ACM/IEEE INTERNATIONAL SYMPOSIUM ON CODE GENERATION AND OPTIMIZATION, CGO 2023, 2023, : 224 - 235