OctoFAS: A Two-Level Fair Scheduler That Increases Fairness in Network-Based Key-Value Storage

Park, Yeohyeon; Park, Junhyeok; Park, Junghwan; Khan, Awais; Kim, Kyeongpyo; Park, Sung-Soon; Kim, Youngjae

doi:10.3390/electronics13030619

OctoFAS: A Two-Level Fair Scheduler That Increases Fairness in Network-Based Key-Value Storage

Journal Article · Thu Feb 01 00:00:00 EST 2024 · Electronics

DOI:https://doi.org/10.3390/electronics13030619· OSTI ID:2301634

Park, Yeohyeon ^[1]; Park, Junhyeok ^[1]; Park, Junghwan ^[1]; Khan, Awais ^[2]; Kim, Kyeongpyo ^[3]; Park, Sung-Soon ^[4]; ^[1]

Sogang Univ. Seoul (Korea, Republic of)
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
GlueSys Co., Ltd., Anyang (Korea, Republic of)
GlueSys Co., Ltd., Anyang (Korea, Republic of); Anyang Univ. (Korea, Republic of)

We identified a fairness problem in a network-based key-value storage system using Intel Storage Performance Development Kit (SPDK) in a multitenant environment. In such an environment, each tenant’s I/O service rate is not fairly guaranteed compared to that of other tenants. To address the fairness problem, we propose OctoFAS, a two-level fair scheduler designed to improve overall throughput and fairness among tenants. The two-level scheduler of OctoFAS consists of (i) inter-core scheduling and (ii) intra-core scheduling. Through inter-core scheduling, OctoFAS addresses the load imbalance problem that is inherent in SPDK on the storage server by dynamically migrating I/O requests from overloaded cores to underloaded cores, thereby increasing overall throughput. Intra-core scheduling prioritizes handling requests from starving tenants over well-fed tenants within core-specific event queues to ensure fair I/O services among multiple tenants. OctoFAS is deployed on a Linux cluster with SPDK. Through extensive evaluations, we found that OctoFAS ensures that the total system throughput remains high and balanced, while enhancing fairness by approximately 10% compared to the baseline, when both scheduling levels operate in a hybrid fashion.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: Ministry of Science and ICT (MSIT); National Research Foundation of Korea (NRF); USDOE Office of Science (SC)

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 2301634

Journal Information:: Electronics, Journal Name: Electronics Journal Issue: 3 Vol. 13; ISSN 2079-9292

Publisher:: MDPICopyright Statement

Country of Publication:: United States

Language:: English

References (25)

SPDK: A Development Kit to Build High Performance Storage Applications Yang, Ziye; Harris, James R.; Walker, Benjamin 2017 IEEE International Conference on Cloud Computing Technology and Science (CloudCom) https://doi.org/10.1109/CloudCom.2017.14	conference	December 2017
MISE: Providing performance predictability and improving fairness in shared main memory systems Subramanian, L.; Seshadri, V. 2013 IEEE 19th International Symposium on High Performance Computer Architecture (HPCA) https://doi.org/10.1109/HPCA.2013.6522356	conference	February 2013
The Blacklisting Memory Scheduler: Achieving high performance and fairness at low cost Subramanian, Lavanya; Lee, Donghyuk; Seshadri, Vivek 2014 IEEE 32nd International Conference on Computer Design (ICCD) https://doi.org/10.1109/ICCD.2014.6974655	conference	October 2014
QWin: Core Allocation for Enforcing Differentiated Tail Latency SLOs at Shared Storage Backend Ma, Liuying; Liu, Zhenqing; Xiong, Jin 2022 IEEE 42nd International Conference on Distributed Computing Systems (ICDCS) https://doi.org/10.1109/ICDCS54860.2022.00109	conference	July 2022
FLIN: Enabling Fairness and Enhancing Performance in Modern NVMe Solid State Drives Tavakkol, Arash; Sadrosadati, Mohammad; Ghose, Saugata 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA) https://doi.org/10.1109/ISCA.2018.00041	conference	June 2018
Fairness Metrics for Multi-Threaded Processors Vandierendonck, Hans; Seznec, Andre IEEE Computer Architecture Letters, Vol. 10, Issue 1 https://doi.org/10.1109/L-CA.2011.1	journal	January 2011
iLSM-SSD: An Intelligent LSM-Tree Based Key-Value SSD for Data Analytics Lee, Chang-Gyu; Kang, Hyeongu; Park, Donggyu 2019 IEEE 27th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS) https://doi.org/10.1109/MASCOTS.2019.00048	conference	October 2019
OCTOKV: An Agile Network-Based Key-Value Storage System with Robust Load Orchestration Park, Yeohyeon; Park, Junhyeok; Khan, Awais 2023 31st International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS) https://doi.org/10.1109/MASCOTS59514.2023.10387663	conference	October 2023
Fair Queuing Memory Systems Nesbit, Kyle J.; Aggarwal, Nidhi; Laudon, James 2006 39th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO'06) https://doi.org/10.1109/MICRO.2006.24	conference	December 2006
Stall-Time Fair Memory Access Scheduling for Chip Multiprocessors Mutlu, Onur; Moscibroda, Thomas 40th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO 2007) https://doi.org/10.1109/MICRO.2007.21	conference	January 2007
High Throughput Disk Scheduling with Fair Bandwidth Distribution Valente, Paolo; Checconi, Fabio IEEE Transactions on Computers, Vol. 59, Issue 9 https://doi.org/10.1109/TC.2010.105	journal	September 2010
Fair-ZNS: Enhancing Fairness in ZNS SSDs through Self-balancing I/O Scheduling Liu, Renping; Tan, Zhenhua; Shen, Yan IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems https://doi.org/10.1109/TCAD.2022.3232997	journal	January 2024
BLISS: Balancing Performance, Fairness and Complexity in Memory Access Scheduling Subramanian, Lavanya; Lee, Donghyuk; Seshadri, Vivek IEEE Transactions on Parallel and Distributed Systems, Vol. 27, Issue 10 https://doi.org/10.1109/TPDS.2016.2526003	journal	October 2016
TridentKV: A Read-Optimized LSM-Tree Based KV Store via Adaptive Indexing and Space-Efficient Partitioning Lu, Kai; Zhao, Nannan; Wan, Jiguang IEEE Transactions on Parallel and Distributed Systems, Vol. 33, Issue 8 https://doi.org/10.1109/TPDS.2021.3118599	journal	August 2022
Aash Kazempour, Vahid; Kamali, Ali; Fedorova, Alexandra ACM SIGPLAN Notices, Vol. 45, Issue 7 https://doi.org/10.1145/1837854.1736011	journal	March 2010
A case for NUMA-aware contention management on multicore systems Blagodurov, Sergey; Zhuravlev, Sergey; Fedorova, Alexandra Proceedings of the 19th international conference on Parallel architectures and compilation techniques https://doi.org/10.1145/1854273.1854350	conference	September 2010
Fairness via Source Throttling Ebrahimi, Eiman; Lee, Chang Joo; Mutlu, Onur ACM Transactions on Computer Systems, Vol. 30, Issue 2 https://doi.org/10.1145/2166879.2166881	journal	April 2012
Traffic management: a holistic approach to memory placement on NUMA systems Dashti, Mohammad; Fedorova, Alexandra; Funston, Justin Proceedings of the eighteenth international conference on Architectural support for programming languages and operating systems - ASPLOS '13 https://doi.org/10.1145/2451116.2451157	conference	January 2013
Flash storage disaggregation Klimovic, Ana; Kozyrakis, Christos; Thereska, Eno Proceedings of the Eleventh European Conference on Computer Systems https://doi.org/10.1145/2901318.2901337	conference	April 2016
Performance Characterization of NVMe-over-Fabrics Storage Disaggregation Guz, Zvika; Li, Harry (Huan); Shayesteh, Anahita ACM Transactions on Storage, Vol. 14, Issue 4 https://doi.org/10.1145/3239563	journal	November 2018
Gimbal Min, Jaehong; Liu, Ming; Chugh, Tapan Proceedings of the 2021 ACM SIGCOMM 2021 Conference https://doi.org/10.1145/3452296.3472940	conference	August 2021
Design of LSM-tree-based Key-value SSDs with Bounded Tails Im, Junsu; Bae, Jinwook; Chung, Chanwoo ACM Transactions on Storage, Vol. 17, Issue 2 https://doi.org/10.1145/3452846	journal	May 2021
Towards low-latency I/O services for mixed workloads using ultra-low latency SSDs Liu, Mingzhe; Liu, Haikun; Ye, Chencheng Proceedings of the 36th ACM International Conference on Supercomputing https://doi.org/10.1145/3524059.3532378	conference	June 2022
Iterator Interface Extended LSM-tree-based KVSSD for Range Queries Lee, Seungjin; Lee, Chang-Gyu; Min, Donghyun Proceedings of the 16th ACM International Conference on Systems and Storage https://doi.org/10.1145/3579370.3594775	conference	June 2023
Load balancing in cloud computing – A hierarchical taxonomical classification Afzal, Shahbaz; Kavitha, G. Journal of Cloud Computing, Vol. 8, Issue 1 https://doi.org/10.1186/s13677-019-0146-7	journal	December 2019

Similar Records

OCTOKV: An Agile Network-Based Key-Value Storage System with Robust Load Orchestration

Conference · Sun Oct 01 00:00:00 EDT 2023 · OSTI ID:2301638

Interactivity vs. fairness in networked linux systems

Journal Article · Sun Dec 31 23:00:00 EST 2006 · OSTI ID:927937

Improving Block-level Efficiency with scsi-mq

Conference · Wed Dec 31 23:00:00 EST 2014 · OSTI ID:1265714

Related Subjects

97 MATHEMATICS AND COMPUTING
I/O scheduling
high-performance I/O
key-value store
storage system

OctoFAS: A Two-Level Fair Scheduler That Increases Fairness in Network-Based Key-Value Storage

Citation Formats

References (25)

Similar Records

Related Subjects