Scalable and quantitative contention generation for performance evaluation on OLTP databases

Abstract

Massive scale of transactions with critical requirements become popular for emerging businesses, especially in E-commerce. One of the most representative applications is the promotional event running on Alibaba’s platform on some special dates, widely expected by global customers. Although we have achieved significant progress in improving the scalability of transactional database systems (OLTP), the presence of contention operations in workloads is still one of the fundamental obstacles to performance improving. The reason is that the overhead of managing conflict transactions with concurrency control mechanisms is proportional to the amount of contentions. As a consequence, generating contented workloads is urgent to evaluate performance of modern OLTP database systems. Though we have kinds of standard benchmarks which provide some ways in simulating contentions, e.g., skew distribution control of transactions, they can not control the generation of contention quantitatively; even worse, the simulation effectiveness of these methods is affected by the scale of data. So in this paper we design a scalable quantitative contention generation method with fine contention granularity control. We conduct a comprehensive set of experiments on popular opensourced DBMSs compared with the latest contention simulation method to demonstrate the effectiveness of our generation work.

Graphical abstract

Keywords

high contention / OLTP database / performance evaluation / database benchmarking

References

[1] Huang G, Cheng X, Wang J, Wang Y, He D, Zhang T, Li F, Wang S, Cao W, Li Q. X-engine: an optimized storage engine for large-scale ecommerce transaction processing. In: Proceedings of 2019 International Conference on Management of Data. 2019, 651– 665

[2] Appuswamy R, Anadiotis A C, Porobic D, Iman M K, Ailamaki A . Analyzing the impact of system architecture on the scalability of OLTP engines for high-contention workloads. Proceedings of the VLDB Endowment, 2017, 11( 2): 121– 134

[3] Harding R, Van Aken D, Pavlo A, Stonebraker M . An evaluation of distributed concurrency control. Proceedings of the VLDB Endowment, 2017, 10( 5): 553– 564

[4] Lan H, Bao Z, Peng Y. A survey on advancing the DBMS query optimizer: cardinality estimation, cost model, and plan enumeration. Data Science and Engineering, 2021, 6(1): 86– 101

[5] Ji Y, Chai Y, Zhou X, Ren L, Qin Y. Smart intra-query fault tolerance for massive parallel processing databases. Data Science And Engineering, 2020, 5(1): 65– 79

[6] Yu X, Bezerra G, Pavlo A. Staring into the abyss: an evaluation of concurrency control with one thousand cores. Proceedings of the VLDB Endowment, 2014, 8(3): 209– 220

[7] Johnson F R. Scalable storage managers for the multicore era. Carnegie Mellon University, Dissertation, 2010

[8] Faleiro J M, Abadi D J. FIT: a distributed database performance tradeoff. IEEE Data Engineering Bulletin, 2015, 38(1): 10– 17

[9] Gurajada A, Gala D, Zhou F, Pathak A, Ma Z F . BTrim: hybrid in-memory database architecture for extreme transaction processing. Proceedings of the VLDB Endowment, 2018, 11( 12): 1889– 1901

[10] Ren K, Faleiro J M, Abadi D J. Design principles for scaling multi-core OLTP under high contention. In: Proceedings of the 2016 International Conference on Management of Data. 2016, 1583– 1598

[11] Zhou N, Zhou X, Zhang X, Du X, Wang S . Reordering transaction execution to boost high-frequency trading applications. Data Science and Engineering, 2017, 2( 4): 301– 315

[12] Chen S, Ailamaki A, Athanassoulis M, Gibbons P B, Johnson R, Pandis I, Stoica R . TPC-E vs. TPC-C: characterizing the new TPC-E benchmark via an I/O comparison study. Acm Sigmod Record, 2011, 39( 3): 5– 10

[13] Cooper B F, Silberstein A, Tam E, Ramakrishnan R, Sears R. Benchmarking cloud serving systems with YCSB. In: Proceedings of the 1st ACM Symposium on Cloud Computing. 2010. 143− 154

[14] Zhang C, Li Y, Zhang R, Qian W, Zhou A. Benchmarking on intensive transaction processing. Frontiers of Computer Science, 2020, 14(5): 145204

[15] Bodik P, Fox A, Franklin M J, Jordan M I, David A. Patterson: characterizing, modeling, and generating workload spikes for stateful services. In: Proceedings of the 1st ACM Symposium on Cloud Computing. 2010, 241− 252

[16] Wolski A. TATP benchmark description.

[17] Stonebraker M. A measure of transaction processing power. In: Stonebraker M, ed. Readings in Database Systems. 2nd ed. San Francisco: Morgan Kaufmann Publishers Inc. 1994, 442− 454

[18] Cahill M J, Röhm U, Fekete A D . Fekete: serializable isolation for snapshot databases. ACM Transactions on Database Systems, 2009, 34( 4): 20

[19] Difallah D E, Pavlo A, Curino C, Cudre-Mauroux P. OLTP-bench: an extensible testbed for benchmarking relational databases. Proceedings of the VLDB Endowment, 2013, 7(4): 277– 288

[20] MySQL database. See Mysql.com website, 2021

[21] PostgreSQL database. See Postgresql.org website, 2021

RIGHTS & PERMISSIONS

Higher Education Press