TY - JOUR
T1 - LADIO
T2 - Leakage-Aware Direct I/O for I/O-Intensive Workloads
AU - Jeong, Ipoom
AU - Lou, Jiaqi
AU - Son, Yongseok
AU - Park, Yongjoo
AU - Yuan, Yifan
AU - Kim, Nam Sung
N1 - This work was supported in part by Intel Transformative Server Architectures (TSA) Center, in part by Samsung Electronics and PRISM, one of the seven centers in JUMP 2.0, in part by DARPA through a Semiconductor Research Corporation (SRC) Program, and in part by theNationalResearch Foundation ofKorea underGrants 2021R1C1C1010861 and 2022R1A4A5034130.
PY - 2023/7/1
Y1 - 2023/7/1
N2 - The advancement in I/O technology has posed an unprecedented demand for high-performance processing on I/O data, leading to the development of Data Direct I/O (DDIO) technology. DDIO improves I/O processing efficiency by directly injecting all inbound I/O data into the last-level cache (LLC) in cooperation with any type of I/O device. Nonetheless, in certain scenarios with more than one I/O applications, DDIO may have sub-optimal performance caused by interference inside the LLC, resulting in the degradation of system performance. Especially, in this paper, we demonstrate that storage I/O on modern high-performance NVMe SSDs hardly benefits from DDIO, sometimes causing inefficient use of the shared LLC due to the 'leaky DMA problem'. To address this problem, we propose LADIO, an adaptive approach that mitigates inter-application interference by dynamically controlling the DDIO functionality and reallocating LLC ways based on the leakage and locality of storage I/O data, respectively. In scenarios with heavy I/O interference, LADIO improves the throughput of network-intensive applications by 20% while maintaining that of storage-intensive applications.
AB - The advancement in I/O technology has posed an unprecedented demand for high-performance processing on I/O data, leading to the development of Data Direct I/O (DDIO) technology. DDIO improves I/O processing efficiency by directly injecting all inbound I/O data into the last-level cache (LLC) in cooperation with any type of I/O device. Nonetheless, in certain scenarios with more than one I/O applications, DDIO may have sub-optimal performance caused by interference inside the LLC, resulting in the degradation of system performance. Especially, in this paper, we demonstrate that storage I/O on modern high-performance NVMe SSDs hardly benefits from DDIO, sometimes causing inefficient use of the shared LLC due to the 'leaky DMA problem'. To address this problem, we propose LADIO, an adaptive approach that mitigates inter-application interference by dynamically controlling the DDIO functionality and reallocating LLC ways based on the leakage and locality of storage I/O data, respectively. In scenarios with heavy I/O interference, LADIO improves the throughput of network-intensive applications by 20% while maintaining that of storage-intensive applications.
KW - Cache
KW - data direct I/O (DDIO)
KW - datacenters
KW - network
KW - storage
UR - http://www.scopus.com/inward/record.url?scp=85164416107&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85164416107&partnerID=8YFLogxK
U2 - 10.1109/LCA.2023.3290427
DO - 10.1109/LCA.2023.3290427
M3 - Article
AN - SCOPUS:85164416107
SN - 1556-6056
VL - 22
SP - 77
EP - 80
JO - IEEE Computer Architecture Letters
JF - IEEE Computer Architecture Letters
IS - 2
ER -