Abstract
In a heterogeneous cluster based on the Raft protocol, in order to solve the problem of slow performance caused by the leader on a slow node, someone proposed ALOR. However, the leader distribution of ALOR is not optimal. In this paper, we propose Workload-driven Adaptive Layout Optimization of Raft groups (WALOR), which changes the leader distribution of ALOR to promote the performance further by more fitting the read-write request ratio of the system’s workload. Our experiments on an actual heterogeneous cluster show that, on average, WALOR improves throughput by 82.96% and 32.42% compared to the even distribution (ED) solution and ALOR, respectively.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ongaro, D, Ousterhout, J.: In search of an understandable consensus algorithm. In: 2014 USENIX Annual Technical Conference (USENIXATC 14), pp. 305–319 (2014)
Ongaro, D.: Consensus: bridging theory and practice. Stanford University (2014)
Wang, Y., Chai, Y., Wang, X.: ALOR: adaptive layout optimization of raft groups for heterogeneous distributed key-Value stores. In: Zhang, F., Zhai, J., Snir, M., Jin, H., Kasahara, H., Valero, M. (eds.) NPC 2018. LNCS, vol. 11276, pp. 13–26. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-05677-3_2
TiKV. https://github.com/tikv/tikv (2022)
Cooper, B.F., et al.: Benchmarking cloud serving systems with YCSB. In: Proceedings of the 1st ACM symposium on Cloud computing, pp. 143–154. ACM (2010)
Lamport, L.: The part-time parliament. ACM Trans. Comput. Syst. (TOCS) 16(2), 133–169 (1998)
Lamport, L.: Paxos made simple. ACM SIGACT News 32(4), 18–25 (2001)
Where can I get Raft? https://raft.github.io/#implementations (2022)
Etcd. https://github.com/etcd-io/etcd (2022)
Corbett, J.C., Dean, J., Epstein, M., et al.: Spanner: google’s globally distributed database. ACM Trans. Comput. Syst. (TOCS) 31(3), 1–22 (2013)
Huang, D., et al.: TiDB: a Raft-based HTAP database. Proc. of the VLDB Endowment 13(12), 3072–3084 (2020)
Cao, W., et al.: POLARDB meets computational storage: efficiently support analytical workloads in cloud-native relational database. In: FAST (2020)
Little, J.D.C.: A proof for the queuing formula: L= \(\lambda \)W. Oper. Res. 9(3), 383–387 (1961)
Little, J.D.C.: OR FORUM-Little’s Law as viewed on its 50th anniversary. Oper. Res. 59(3), 536–549 (2011)
Liu, G., Wang, S., Bao, Y.: SEER: a time prediction model for CNNs from GPU kernel’s view. In: 2021 30th International Conference on Parallel Architectures and Compilation Techniques (PACT), pp. 173–185. In: IEEE (2021)
Volkov, V.: A microbenchmark to study GPU performance models. ACM SIGPLAN Not. 53(1), 421–422 (2018)
go-ycsb. https://github.com/pingcap/go-ycsb (2022)
Wang, C., et al.: Apus: fast and scalable PAXOS on RDMA In: Proceedings of the 2017 Symposium on Cloud Computing, pp. 94–107. ACM (2017)
Aguilera, M.K., et al.: Microsecond consensus for microsecond applications. In: Operating Systems Design and Implementation. USENIX ASSOC (2020)
Cao, W., Liu, Z., Wang, P., et al.: PolarFS: an ultra-low latency and failure resilient distributed file system for shared storage cloud database. Proc. VLDB Endowment 11(12), 1849–1862 (2018)
Sakic, E., Kellerer, W.: Response time and availability study of RAFT consensus in distributed SDN control plane. IEEE Trans. Netw. Serv. Manag. 15(1), 304–318 (2017)
Zhang, Y., et al.: When raft meets SDN: how to elect a leader and reach consensus in an unruly network. In: Proceedings of the First Asia-Pacific Workshop on Networking, pp. 1–7. ACM (2017)
Kim, T., et al.: Load balancing on distributed datastore in opendaylight SDN controller cluster. In: 2017 IEEE Conference on Network Softwarization (NetSoft), pp. 1–3. IEEE (2017)
Copeland, C., Zhong, H.: Tangaroa: a byzantine fault tolerant raft (2016)
Dadheech, P., et, al.: Performance improvement of heterogeneous cluster of big data using query optimization and mapreduce. In: International Conference on Information Management and Machine Intelligence (ICIMMI 2019) (2020)
Yuan, Y., et al.: A distributed in-memory key-value store system on heterogeneous CPU-GPU cluster. VLDB J. 26(5), 729–750 (2017)
Kwon, Y., et al.: Strata: a cross media file system. In: The 26th Symposium (2017)
Kakoulli, E., Herodotou, H.: OctopusFS: a distributed file system with tiered storage management. In: The 2017 ACM International Conference. ACM (2017)
Acknowledgement
This work is supported by the National Key Research and Development Program of China (No. 2019YFE0198600), National Natural Science Foundation of China (No. 61972402, 61972275, and 61732014).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 IFIP International Federation for Information Processing
About this paper
Cite this paper
Wang, Y., Chai, Y., Zhang, Q. (2022). WALOR: Workload-Driven Adaptive Layout Optimization of Raft Groups for Heterogeneous Distributed Key-Value Stores. In: Liu, S., Wei, X. (eds) Network and Parallel Computing. NPC 2022. Lecture Notes in Computer Science, vol 13615. Springer, Cham. https://doi.org/10.1007/978-3-031-21395-3_27
Download citation
DOI: https://doi.org/10.1007/978-3-031-21395-3_27
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-21394-6
Online ISBN: 978-3-031-21395-3
eBook Packages: Computer ScienceComputer Science (R0)