TY - GEN
T1 - STYX
T2 - 2023 USENIX Annual Technical Conference, ATC 2023
AU - Ji, Houxiang
AU - Mansi, Mark
AU - Sun, Yan
AU - Yuan, Yifan
AU - Huang, Jinghan
AU - Kuper, Reese
AU - Swift, Michael M.
AU - Kim, Nam Sung
N1 - We thank Jiacheng Ma and Ipoom Jeong for their technical discussion and support. This work was supported in part by Samsung Electronics, the IBM-Illinois Discovery Accelerator Institute and PRISM, one of the seven centers in JUMP 2.0, a Semiconductor Research Corporation (SRC) program sponsored by DARPA. Nam Sung Kim has a financial interest in Samsung Electronics and NeuroRealityVision.
PY - 2023
Y1 - 2023
N2 - 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 99thSpercentile latency of co-running applications, compared to a system without STYX, while preserving the benefits of these kernel features.
AB - 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 99thSpercentile latency of co-running applications, compared to a system without STYX, while preserving the benefits of these kernel features.
UR - http://www.scopus.com/inward/record.url?scp=85177594554&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85177594554&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85177594554
T3 - Proceedings of the 2023 USENIX Annual Technical Conference, ATC 2023
SP - 619
EP - 633
BT - Proceedings of the 2023 USENIX Annual Technical Conference, ATC 2023
PB - USENIX Association
Y2 - 10 July 2023 through 12 July 2023
ER -