Abstract
To improve the performance for high-contention workloads, modern main-memory database systems seek to design efficient lock managers. However, OLTP engines adopt the classic FCFS strategy to process operations, where the generated execution order does not take current and future conflicts into consideration. In this case, lock dependencies will happen more frequently and thus resulting in high transaction waiting time, referred to as lock thrashing, which is proved to be the main bottleneck of lock-based concurrency control mechanisms. In this paper, we present a transaction scheduler that generates efficient execution order to alleviate the lock thrashing issue. To proactively resolve conflicts, LOTAS predicts which data will be accessed by following operations through building Markov-based prediction graphs. Then LOTAS uses the information to schedule transactions by judging whether a transaction needs to be deferred to acquire locks. Experimental results demonstrate that LOTAS can significantly reduce the lock waiting time and improves the throughput up to 4.8x than the classic FCFS strategy under highly contended workloads.
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
Dbx1000. https://github.com/yxymit/DBx1000
Bernstein, P.A., Newcomer, E.: Principles of Transaction Processing for Systems Professionals. Morgan Kaufmann, Burlington (1996)
Ding, B., Kot, L., Gehrke, J.: Improving optimistic concurrency control through transaction batching and operation reordering. PVLDB 12(2), 169–182 (2018)
Faleiro, J.M., Abadi, D., Hellerstein, J.M.: High performance transactions via early write visibility. PVLDB 10(5), 613–624 (2017)
Ganapathi, A., Kuno, H.A., et al.: Predicting multiple metrics for queries: better decisions enabled by machine learning. In: ICDE, pp. 592–603 (2009)
Guo, J., Chu, J., Cai, P., Zhou, M., Zhou, A.: Low-overhead paxos replication. Data Sci. Eng. 2(2), 169–177 (2017). https://doi.org/10.1007/s41019-017-0039-z
Gupta, C., Mehta, A., Dayal, U.: PQR: predicting query execution times for autonomous workload management. In: ICAC, pp. 13–22 (2008)
Haritsa, J.R., Carey, M.J., Livny, M.: Dynamic real-time optimistic concurrency control. In: RTSS, pp. 94–103 (1990)
Harizopoulos, S., Abadi, D.J., Madden, S., Stonebraker, M.: OLTP through the looking glass, and what we found there. In: SIGMOD, pp. 981–992 (2008)
Huang, J., Mozafari, B., et al.: A top-down approach to achieving performance predictability in database systems. In: SIGMOD, pp. 745–758 (2017)
Jung, H., Han, H., Fekete, A.D., Heiser, G., Yeom, H.Y.: A scalable lock manager for multicores. In: SIGMOD, pp. 73–84 (2013)
Ma, L., Aken, D.V., Hefny, A., et al.: Query-based workload forecasting for self-driving database management systems. In: SIGMOD, pp. 631–645 (2018)
McWherter, D.T., Schroeder, B., Ailamaki, A., et al.: Priority mechanisms for OLTP and transactional web applications. In: ICDE, pp. 535–546 (2004)
McWherter, D.T., Schroeder, B., Ailamaki, A., et al.: Improving preemptive prioritization via statistical characterization of OLTP locking. In: ICDE, pp. 446–457 (2005)
Mu, S., Cui, Y., Zhang, Y., Lloyd, W., Li, J.: Extracting more concurrency from distributed transactions. In: OSDI, pp. 479–494 (2014)
Pavlo, A.: What are we doing with our lives?: nobody cares about our concurrency control research. In: SIGMOD, p. 3 (2017)
Pavlo, A., Jones, E.P.C., Zdonik, S.B.: On predictive modeling for optimizing transaction execution in parallel OLTP systems. PVLDB 5(2), 85–96 (2011)
Prasaad, G., Cheung, A., Suciu, D.: Improving high contention OLTP performance via transaction scheduling. CoRR, abs/1810.01997 (2018)
Ren, K., Faleiro, J.M., Abadi, D.J.: Design principles for scaling multi-core OLTP under high contention. In: SIGMOD, pp. 1583–1598 (2016)
Ren, K., Thomson, A., Abadi, D.J.: Lightweight locking for main memory database systems. PVLDB 6(2), 145–156 (2012)
Sapia, C.: PROMISE: predicting query behavior to enable predictive caching strategies for OLAP systems. In: Kambayashi, Y., Mohania, M., Tjoa, A.M. (eds.) DaWaK 2000. LNCS, vol. 1874, pp. 224–233. Springer, Heidelberg (2000). https://doi.org/10.1007/3-540-44466-1_22
Shasha, D.E., Llirbat, F., Simon, E., et al.: Transaction chopping: algorithms and performance studies. ACM Trans. Database Syst. 20(3), 325–363 (1995)
Thomson, A., Diamond, T., Weng, S., et al.: Calvin: fast distributed transactions for partitioned database systems. In: SIGMOD, pp. 1–12 (2012)
Tian, B., Huang, J., Mozafari, B., Schoenebeck, G.: Contention-aware lock scheduling for transactional databases. PVLDB 11(5), 648–662 (2018)
Wang, T., Kimura, H.: Mostly-optimistic concurrency control for highly contended dynamic workloads on a thousand cores. PVLDB 10(2), 49–60 (2016)
Wang, Z., Mu, S., Cui, Y., Yi, H., Chen, H., Li, J.: Scaling multicore databases via constrained parallel execution. In: SIGMOD, pp. 1643–1658 (2016)
Xie, C., Su, C., Littley, C., Alvisi, L., Kapritsos, M., Wang, Y.: High-performance ACID via modular concurrency control. In: SOSP, pp. 279–294 (2015)
Yan, C., Cheung, A.: Leveraging lock contention to improve OLTP application performance. PVLDB 9(5), 444–455 (2016)
Yu, X., Bezerra, G., Pavlo, A., et al.: Staring into the abyss: an evaluation of concurrency control with one thousand cores. PVLDB 8(3), 209–220 (2014)
Zhang, T., Tomasic, A., Sheng, Y., Pavlo, A.: Performance of OLTP via intelligent scheduling. In: ICDE, pp. 1288–1291 (2018)
Acknowledgement
This work is partially supported by National Key R&D Program of China (2018YFB1003404), and NSFC under grant numbers 61972149, and ECNU Academic Innovation Promotion Program for Excellent Doctoral Students. The corresponding author is Peng Cai. We thank anonymous reviewers for their very helpful comments.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Wang, D., Cai, P., Qian, W., Zhou, A. (2020). Predictive Transaction Scheduling for Alleviating Lock Thrashing. In: Nah, Y., Cui, B., Lee, SW., Yu, J.X., Moon, YS., Whang, S.E. (eds) Database Systems for Advanced Applications. DASFAA 2020. Lecture Notes in Computer Science(), vol 12112. Springer, Cham. https://doi.org/10.1007/978-3-030-59410-7_9
Download citation
DOI: https://doi.org/10.1007/978-3-030-59410-7_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-59409-1
Online ISBN: 978-3-030-59410-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)