research-article

Scaling up concurrent main-memory column-store scans: towards adaptive NUMA-aware data and task placement

Editors: Chen Li, Volker Markl Authors:

Iraklis Psaroudakis,

Tobias Scheuer,

Abdelkader Sellami,

Anastasia AilamakiAuthors Info & Claims

Proceedings of the VLDB Endowment, Volume 8, Issue 12

Pages 1442 - 1453

https://doi.org/10.14778/2824032.2824043

Published: 01 August 2015 Publication History

Abstract

Main-memory column-stores are called to efficiently use modern non-uniform memory access (NUMA) architectures to service concurrent clients on big data. The efficient usage of NUMA architectures depends on the data placement and scheduling strategy of the column-store. Most column-stores choose a static strategy that involves partitioning all data across the NUMA architecture, and employing a stealing-based task scheduler. In this paper, we implement different strategies for data placement and task scheduling for the case of concurrent scans. We compare these strategies with an extensive sensitivity analysis. Our most significant findings include that unnecessary partitioning can hurt throughput by up to 70%, and that stealing memory-intensive tasks can hurt throughput by up to 58%. Based on our analysis, we envision a design that adapts the data placement and task scheduling strategy to the workload.

References

[1]

Intel Xeon Processor E7 Family Uncore Performance Monitoring, 2011. http://www.intel.com/.

[2]

Intel Xeon Processor E7 v2 2800/4800/8800 - Datasheet - Vol. 2, 2014. http://www.intel.com/.

[3]

SAP BW Enhanced Mixed Load benchmark, 2015. http://global.sap.com/campaigns/benchmark/.

[4]

SAP HANA Platform SQL and System Views Reference, 2015. http://help.sap.com.

[5]

A. Ailamaki et al. How to stop under-utilization and love multicores. In SIGMOD, pp. 189--192, 2014.

[6]

M.-C. Albutiu et al. Massively Parallel Sort-Merge Joins in Main Memory Multi-Core Database Systems. VLDB, 5(10):1064--1075, 2012.

[7]

D. Alistarh et al. The SprayList: A Scalable Relaxed Priority Queue. In PPoPP, pp. 11--20, 2015.

[8]

C. Balkesen et al. Multi-Core, Main-Memory Joins: Sort vs. Hash Revisited. VLDB, 7(1):85--96, 2013.

[9]

T. Becker et al. Searching for the Best System Configuration to Fit Your BW?, Nov. 2012. http://scn.sap.com/docs/DOC-33705.

[10]

S. Blagodurov et al. A Case for NUMA-aware Contention Management on Multicore Systems. In USENIXATC, pp. 557--558, 2011.

[11]

W. Chen et al. Architecting and Deploying DB2 with BLU Acceleration. IBM Redbooks, 2014.

[12]

TPC-H Benchmark Rev. 2.17.1. http://www.tpc.org.

[13]

M. Dashti et al. Traffic Management: A Holistic Approach to Memory Placement on NUMA Systems. In ASPLOS, pp. 381--394, 2013.

[14]

F. Färber et al. The SAP HANA Database -- An Architecture Overview. IEEE Data Eng. Bull., 35(1):28--33, 2012.

[15]

J. Giceva et al. Deployment of Query Plans on Multicores. VLDB, 8(3):233--244, 2014.

[16]

T. Gubner. Achieving Many-Core Scalability in Vectorwise. Master's thesis, TU Ilmenau, 2014.

[17]

T. Kissinger et al. ERIS: A NUMA-Aware In-Memory Storage Engine for Analytical Workloads. In ADMS, pp. 74--85, 2014.

[18]

T. Lahiri et al. Oracle Database In-Memory: A Dual Format In-Memory Database. In ICDE, 2015.

[19]

C. Lameter et al. NUMA (Non-Uniform Memory Access): An Overview. ACM Queue, 11(7):40, 2013.

[20]

H. Lang et al. Massively Parallel NUMA-aware Hash Joins. In IMDM, 2013.

[21]

P.-A. Larson et al. Enhancements to SQL server column stores. In SIGMOD, pp. 1159--1168, 2013.

[22]

T. Legler et al. Data Mining with the SAP NetWeaver BI Accelerator. In VLDB, pp. 1059--1068, 2006.

[23]

V. Leis et al. Morsel-Driven Parallelism: A NUMA-Aware Query Evaluation Framework for the Many-Core Age. In SIGMOD, pp. 743--754, 2014.

[24]

C. Lemke et al. Speeding up queries in column stores: a case for compression. In DaWaK, pp. 117--129, 2010.

[25]

Y. Li et al. NUMA-aware algorithms: the case of data shuffling. In CIDR, 2013.

[26]

S. Idreos et al. MonetDB: Two decades of research in column-oriented database architectures. Data Engineering, page 40, 2012.

[27]

D. Porobic et al. ATraPos: Adaptive transaction processing on hardware Islands. In ICDE, pp. 688--699, 2014.

[28]

I. Psaroudakis et al. Task Scheduling for Highly Concurrent Analytical and Transactional Main-Memory Workloads. In ADMS, pp. 36--45, 2013.

[29]

I. Psaroudakis et al. Scaling Up Mixed Workloads: A Battle of Data Freshness, Flexibility, and Scheduling. In TPCTC, pp. 97--112, 2015.

[30]

V. Raman et al. DB2 with BLU Acceleration: So much more than just a column store. VLDB, 6(11):1080--1091, 2013.

[31]

P. Russom. Best Practices Report: High-Performance Data Warehousing, 2012. http://tdwi.org/.

[32]

M. Stonebraker et al. The end of an architectural era (it's time for a complete rewrite). In VLDB, pp. 1150--1160, 2007.

[33]

T. Willhalm et al. SIMD-Scan: Ultra Fast in-Memory Table Scan using on-Chip Vector Processing Units. In VLDB, volume 2, pp. 385--394, 2009.

[34]

T. Willhalm et al. Vectorizing database column scans with complex predicates. In ADMS, pp. 1--12, 2013.

[35]

F. Wolf et al. Extending Database Task Schedulers for Multi-threaded Application Code. In SSDBM, 2015, to appear.

[36]

M. Zukowski et al. Vectorwise: Beyond Column Stores. IEEE Data Eng. Bull., 35(1):21--27, 2012.

Cited By

Wang YJia WJiang DXiong J(2023)A Survey of Non-Volatile Main Memory File SystemsJournal of Computer Science and Technology10.1007/s11390-023-1054-338:2(348-372)Online publication date: 30-Mar-2023
https://dl.acm.org/doi/10.1007/s11390-023-1054-3
Sabek IUkyab TKraska TIves ZBonifati AEl Abbadi A(2022)LSched: A Workload-Aware Learned Query Scheduler for Analytical Database SystemsProceedings of the 2022 International Conference on Management of Data10.1145/3514221.3526158(1228-1242)Online publication date: 10-Jun-2022
https://dl.acm.org/doi/10.1145/3514221.3526158
Bang TMay NPetrov IBinnig C(2022)The full story of 1000 coresThe VLDB Journal — The International Journal on Very Large Data Bases10.1007/s00778-022-00742-431:6(1185-1213)Online publication date: 29-Apr-2022
https://dl.acm.org/doi/10.1007/s00778-022-00742-4
Show More Cited By

Scaling up concurrent main-memory column-store scans: towards adaptive NUMA-aware data and task placement

Recommendations

Store Atomicity for Transactional Memory

We extend the notion of Store Atomicity [Arvind and Jan-Willem Maessen. Memory model = instruction reordering + store atomicity. In ISCA '06: Proceedings of the 33rd annual International Symposium on Computer Architecture, 2006] to a system with atomic ...
Redesign the Memory Allocator for Non-Volatile Main Memory
Special Issue on Hardware and Algorithms for Learning On-a-chip and Special Issue on Alternative Computing Systems

The non-volatile memory (NVM) has the merits of byte-addressability, fast speed, persistency and low power consumption, which make it attractive to be used as main memory. Commonly, user process dynamically acquires memory through memory allocators. ...
Low overhead concurrency control for partitioned main memory databases
SIGMOD '10: Proceedings of the 2010 ACM SIGMOD International Conference on Management of data

Database partitioning is a technique for improving the performance of distributed OLTP databases, since "single partition" transactions that access data on one partition do not need coordination with other partitions. For workloads that are amenable to ...

Comments

Information & Contributors

Information

Published In

cover image Proceedings of the VLDB Endowment

Proceedings of the VLDB Endowment Volume 8, Issue 12

Proceedings of the 41st International Conference on Very Large Data Bases, Kohala Coast, Hawaii

August 2015

728 pages

ISSN:2150-8097

Editors:
Chen Li
University of California, Irvine
,
Volker Markl
TU Berlin

Issue’s Table of Contents

Publisher

VLDB Endowment

Publication History

Published: 01 August 2015

Published in PVLDB Volume 8, Issue 12

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

21
Total Citations
View Citations
265
Total Downloads

Downloads (Last 12 months)17
Downloads (Last 6 weeks)3

Reflects downloads up to 27 Dec 2024

Other Metrics

View Author Metrics

Citations

Cited By

Wang YJia WJiang DXiong J(2023)A Survey of Non-Volatile Main Memory File SystemsJournal of Computer Science and Technology10.1007/s11390-023-1054-338:2(348-372)Online publication date: 30-Mar-2023
https://dl.acm.org/doi/10.1007/s11390-023-1054-3
Sabek IUkyab TKraska TIves ZBonifati AEl Abbadi A(2022)LSched: A Workload-Aware Learned Query Scheduler for Analytical Database SystemsProceedings of the 2022 International Conference on Management of Data10.1145/3514221.3526158(1228-1242)Online publication date: 10-Jun-2022
https://dl.acm.org/doi/10.1145/3514221.3526158
Bang TMay NPetrov IBinnig C(2022)The full story of 1000 coresThe VLDB Journal — The International Journal on Very Large Data Bases10.1007/s00778-022-00742-431:6(1185-1213)Online publication date: 29-Apr-2022
https://dl.acm.org/doi/10.1007/s00778-022-00742-4
Liu GChen LChen S(2022)Zen+: a robust NUMA-aware OLTP engine optimized for non-volatile main memoryThe VLDB Journal — The International Journal on Very Large Data Bases10.1007/s00778-022-00737-132:1(123-148)Online publication date: 6-Apr-2022
https://dl.acm.org/doi/10.1007/s00778-022-00737-1
Muthukrishnan HLustig DNellans DWenisch T(2021)GPS: A Global Publish-Subscribe Model for Multi-GPU Memory ManagementMICRO-54: 54th Annual IEEE/ACM International Symposium on Microarchitecture10.1145/3466752.3480088(46-58)Online publication date: 18-Oct-2021
https://dl.acm.org/doi/10.1145/3466752.3480088
Daase BBollmeier LBenson LRabl TLi GLi ZIdreos SSrivastava D(2021)Maximizing Persistent Memory Bandwidth Utilization for OLAP WorkloadsProceedings of the 2021 International Conference on Management of Data10.1145/3448016.3457292(339-351)Online publication date: 9-Jun-2021
https://dl.acm.org/doi/10.1145/3448016.3457292
Wagner BKohn ANeumann TLi GLi ZIdreos SSrivastava D(2021)Self-Tuning Query Scheduling for Analytical WorkloadsProceedings of the 2021 International Conference on Management of Data10.1145/3448016.3457260(1879-1891)Online publication date: 9-Jun-2021
https://dl.acm.org/doi/10.1145/3448016.3457260
Muthukrishnan HNellans DLustig DFessler JWenisch TMartínez JDuato JJohn L(2021)Efficient multi-GPU shared memory via automatic optimization of fine-grained transfersProceedings of the 48th Annual International Symposium on Computer Architecture10.1109/ISCA52012.2021.00020(139-152)Online publication date: 14-Jun-2021
https://dl.acm.org/doi/10.1109/ISCA52012.2021.00020
Goda KHayamizu YYamada HKitsuregawa M(2020)Out-of-order execution of database queriesProceedings of the VLDB Endowment10.14778/3415478.341557113:12(3489-3501)Online publication date: 14-Sep-2020
https://dl.acm.org/doi/10.14778/3415478.3415571
Bang TMay NPetrov IBinnig CPorobic DNeumann T(2020)The tale of 1000 CoresProceedings of the 16th International Workshop on Data Management on New Hardware10.1145/3399666.3399910(1-9)Online publication date: 15-Jun-2020
https://dl.acm.org/doi/10.1145/3399666.3399910
Show More Cited By

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Article

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Issue’s Table of Contents