Communication Steps for Parallel Query Processing

Journal of the ACM - Tập 64 Số 6 - Trang 1-58 - 2017
Paul Beame1, Paraschos Koutris1, Dan Suciu1
1University of Washington

Tóm tắt

We study the problem of computing conjunctive queries over large databases on parallel architectures without shared storage. Using the structure of such a query q and the skew in the data, we study tradeoffs between the number of processors, the number of rounds of communication, and the per-processor load —the number of bits each processor can send or can receive in a single round—that are required to compute q . Since each processor must store its received bits, the load is at most the number of bits of storage per processor.

When the data are free of skew, we obtain essentially tight upper and lower bounds for one round algorithms, and we show how the bounds degrade when there is skew in the data. In the case of skewed data, we show how to improve the algorithms when approximate degrees of the (necessarily small number of) heavy-hitter elements are available, obtaining essentially optimal algorithms for queries such as skewed simple joins and skewed triangle join queries.

For queries that we identify as treelike , we also prove nearly matching upper and lower bounds for multi-round algorithms for a natural class of skew-free databases. One consequence of these latter lower bounds is that for any ε > 0, using p processors to compute the connected components of a graph, or to output the path, if any, between a specified pair of vertices of a graph with m edges and per-processor load that is O ( m / p 1−ε ) requires Ω(log p ) rounds of communication.

Our upper bounds are given by simple structured algorithms using MapReduce. Our one-round lower bounds are proved in a very general model, which we call the Massively Parallel Communication (MPC) model, that allows processors to communicate arbitrary bits. Our multi-round lower bounds apply in a restricted version of the MPC model in which processors in subsequent rounds after the first communication round are only allowed to send tuples.

Từ khóa


Tài liệu tham khảo

Foto N. Afrati , Manas R. Joglekar , Christopher Ré , Semih Salihoglu , and Jeffrey D. Ullman . 2017. GYM: A multiround distributed join algorithm . In Proceedings of the 20th International Conference on Database Theory (ICDT’17) . 4:1--4:18. DOI:http://dx.doi.org/10.4230/LIPIcs.ICDT. 2017 .4 10.4230/LIPIcs.ICDT.2017.4 Foto N. Afrati, Manas R. Joglekar, Christopher Ré, Semih Salihoglu, and Jeffrey D. Ullman. 2017. GYM: A multiround distributed join algorithm. In Proceedings of the 20th International Conference on Database Theory (ICDT’17). 4:1--4:18. DOI:http://dx.doi.org/10.4230/LIPIcs.ICDT.2017.4

F. N. Afrati A. D. Sarma S. Salihoglu and J. D. Ullman. 2012. Upper and lower bounds on the cost of a map-reduce computation. CoRR abs/1206.4377 (2012). F. N. Afrati A. D. Sarma S. Salihoglu and J. D. Ullman. 2012. Upper and lower bounds on the cost of a map-reduce computation. CoRR abs/1206.4377 (2012).

10.1145/1739041.1739056

10.1006/jcss.1997.1545

10.1109/FOCS.2008.43

10.1145/2463664.2465224

10.1145/2594538.2594558

Johann Brault-Baron . 2016. Hypergraph acyclicity revisited. ACM Comput. Surv. 49, 3 ( 2016 ), 54:1--54:26. http://doi.acm.org/10.1145/2983573 Johann Brault-Baron. 2016. Hypergraph acyclicity revisited. ACM Comput. Surv. 49, 3 (2016), 54:1--54:26. http://doi.acm.org/10.1145/2983573

S. Chaudhuri. 2012. What next?: A half-dozen data management research goals for big data and the cloud. In PODS. 1--4. S. Chaudhuri. 2012. What next?: A half-dozen data management research goals for big data and the cloud. In PODS. 1--4.

10.1145/2723372.2750545

10.1145/240455.240477

J. Dean and S. Ghemawat. 2004. MapReduce: Simplified data processing on large clusters. In OSDI. 137--150. J. Dean and S. Ghemawat. 2004. MapReduce: Simplified data processing on large clusters. In OSDI. 137--150.

EMC Corporation. Data Science Revealed: A Data-Driven Glimpse into the Burgeoning New Field. Retrieved from http://www.emc.com/collateral/about/news/emc-data-science-study-wp.pdf. EMC Corporation. Data Science Revealed: A Data-Driven Glimpse into the Burgeoning New Field. Retrieved from http://www.emc.com/collateral/about/news/emc-data-science-study-wp.pdf.

10.1145/1824777.1824786

10.2307/4145187

A. Gál and P. Gopalan. 2007. Lower bounds on streaming algorithms for approximating the length of the longest increasing subsequence. In FOCS. 294--304. A. Gál and P. Gopalan. 2007. Lower bounds on streaming algorithms for approximating the length of the longest increasing subsequence. In FOCS. 294--304.

10.1016/0743-1066(92)90048-8

10.1007/978-3-642-25591-5_39

10.1145/1568318.1568320

10.1145/1109557.1109590

Sudipto Guha and Zhiyi Huang . 2009. Revisiting the direct sum theorem and space lower bounds in random order streams . In ICALP (LNCS) , Vol. 5555 . Springer , 513--524. Sudipto Guha and Zhiyi Huang. 2009. Revisiting the direct sum theorem and space lower bounds in random order streams. In ICALP (LNCS), Vol. 5555. Springer, 513--524.

10.1145/2588555.2594530

10.1007/978-3-642-15369-3_46

10.1137/1.9781611973075.76

10.1145/3034786.3034788

Paraschos Koutris , Paul Beame , and Dan Suciu . 2016 . Worst-case optimal algorithms for parallel query processing . In Proceedings of the 19th International Conference on Database Theory (ICDT’16) . 8:1--8:18. DOI:http://dx.doi.org/10.4230/LIPIcs.ICDT.2016.8 10.4230/LIPIcs.ICDT.2016.8 Paraschos Koutris, Paul Beame, and Dan Suciu. 2016. Worst-case optimal algorithms for parallel query processing. In Proceedings of the 19th International Conference on Database Theory (ICDT’16). 8:1--8:18. DOI:http://dx.doi.org/10.4230/LIPIcs.ICDT.2016.8

10.1145/1989284.1989310

10.1016/S0065-2458(08)60342-3

10.1145/2213836.2213840

10.14778/1920841.1920886

H. Q. Ngo E. Porat C. Ré and A. Rudra. 2012. Worst-case optimal join algorithms: (Extended abstract). In PODS. 37--48. H. Q. Ngo E. Porat C. Ré and A. Rudra. 2012. Worst-case optimal join algorithms: (Extended abstract). In PODS. 37--48.

10.1145/1376616.1376726

10.1145/2304576.2304607

Spark. Apache Spark. http://spark.apache.org/. Spark. Apache Spark. http://spark.apache.org/.

Siddharth Suri and Sergei Vassilvitskii. 2011. Counting triangles and the curse of the last reducer. In WWW. 607--614. Siddharth Suri and Sergei Vassilvitskii. 2011. Counting triangles and the curse of the last reducer. In WWW. 607--614.

10.14778/1687553.1687609

10.1145/31846.32978

10.1145/2331042.2331053

10.1145/79173.79181

10.14778/3007263.3007307

A. C. Yao. 1977. Probabilistic computations: Toward a unified measure of complexity. In FOCS. 222--227. A. C. Yao. 1977. Probabilistic computations: Toward a unified measure of complexity. In FOCS. 222--227.

Thông tin
Thông tin xuất bản

Journal of the ACM

Tập 64 Số 6

1-58

Thông tin tác giả