Performance prediction for concurrent database workloads

Jennie Duggan*, Ugur Cetintemel, Olga Papaemmanouil, Eli Upfal

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

91 Citations (Scopus)

Abstract

Current trends in data management systems, such as cloud and multi-tenant databases, are leading to data processing environments that concurrently execute heterogeneous query workloads. At the same time, these systems need to satisfy diverse performance expectations. In these newly-emerging settings, avoiding potential Quality-of-Service (QoS) violations heavily relies on performance predictability, i.e., the ability to estimate the impact of concurrent query execution on the performance of individual queries in a continuously evolving workload. This paper presents a modeling approach to estimate the impact of concurrency on query performance for analytical workloads. Our solution relies on the analysis of query behavior in isolation, pairwise query interactions and sampling techniques to predict resource contention under various query mixes and concurrency levels. We introduce a simple yet powerful metric that accurately captures the joint effects of disk and memory contention on query performance in a single value. We also discuss predicting the execution behavior of a time-varying query workload through query-interaction timelines, i.e., a fine-grained estimation of the time segments during which discrete mixes will be executed concurrently. Our experimental evaluation on top of PostgreSQL/TPC-H demonstrates that our models can provide query latency predictions within approximately 20% of the actual values in the average case.

Original languageEnglish (US)
Title of host publicationProceedings of SIGMOD 2011 and PODS 2011
Pages337-348
Number of pages12
DOIs
StatePublished - Jul 11 2011
Event2011 ACM SIGMOD and 30th PODS 2011 Conference - Athens, Greece
Duration: Jun 12 2011Jun 16 2011

Publication series

NameProceedings of the ACM SIGMOD International Conference on Management of Data
ISSN (Print)0730-8078

Other

Other2011 ACM SIGMOD and 30th PODS 2011 Conference
CountryGreece
CityAthens
Period6/12/116/16/11

Fingerprint

Information management
Quality of service
Sampling
Data storage equipment

Keywords

  • concurrency
  • query performance prediction

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this

Duggan, J., Cetintemel, U., Papaemmanouil, O., & Upfal, E. (2011). Performance prediction for concurrent database workloads. In Proceedings of SIGMOD 2011 and PODS 2011 (pp. 337-348). (Proceedings of the ACM SIGMOD International Conference on Management of Data). https://doi.org/10.1145/1989323.1989359
Duggan, Jennie ; Cetintemel, Ugur ; Papaemmanouil, Olga ; Upfal, Eli. / Performance prediction for concurrent database workloads. Proceedings of SIGMOD 2011 and PODS 2011. 2011. pp. 337-348 (Proceedings of the ACM SIGMOD International Conference on Management of Data).
@inproceedings{004e43a5640f4d3685d0348844a4f036,
title = "Performance prediction for concurrent database workloads",
abstract = "Current trends in data management systems, such as cloud and multi-tenant databases, are leading to data processing environments that concurrently execute heterogeneous query workloads. At the same time, these systems need to satisfy diverse performance expectations. In these newly-emerging settings, avoiding potential Quality-of-Service (QoS) violations heavily relies on performance predictability, i.e., the ability to estimate the impact of concurrent query execution on the performance of individual queries in a continuously evolving workload. This paper presents a modeling approach to estimate the impact of concurrency on query performance for analytical workloads. Our solution relies on the analysis of query behavior in isolation, pairwise query interactions and sampling techniques to predict resource contention under various query mixes and concurrency levels. We introduce a simple yet powerful metric that accurately captures the joint effects of disk and memory contention on query performance in a single value. We also discuss predicting the execution behavior of a time-varying query workload through query-interaction timelines, i.e., a fine-grained estimation of the time segments during which discrete mixes will be executed concurrently. Our experimental evaluation on top of PostgreSQL/TPC-H demonstrates that our models can provide query latency predictions within approximately 20{\%} of the actual values in the average case.",
keywords = "concurrency, query performance prediction",
author = "Jennie Duggan and Ugur Cetintemel and Olga Papaemmanouil and Eli Upfal",
year = "2011",
month = "7",
day = "11",
doi = "10.1145/1989323.1989359",
language = "English (US)",
isbn = "9781450306614",
series = "Proceedings of the ACM SIGMOD International Conference on Management of Data",
pages = "337--348",
booktitle = "Proceedings of SIGMOD 2011 and PODS 2011",

}

Duggan, J, Cetintemel, U, Papaemmanouil, O & Upfal, E 2011, Performance prediction for concurrent database workloads. in Proceedings of SIGMOD 2011 and PODS 2011. Proceedings of the ACM SIGMOD International Conference on Management of Data, pp. 337-348, 2011 ACM SIGMOD and 30th PODS 2011 Conference, Athens, Greece, 6/12/11. https://doi.org/10.1145/1989323.1989359

Performance prediction for concurrent database workloads. / Duggan, Jennie; Cetintemel, Ugur; Papaemmanouil, Olga; Upfal, Eli.

Proceedings of SIGMOD 2011 and PODS 2011. 2011. p. 337-348 (Proceedings of the ACM SIGMOD International Conference on Management of Data).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Performance prediction for concurrent database workloads

AU - Duggan, Jennie

AU - Cetintemel, Ugur

AU - Papaemmanouil, Olga

AU - Upfal, Eli

PY - 2011/7/11

Y1 - 2011/7/11

N2 - Current trends in data management systems, such as cloud and multi-tenant databases, are leading to data processing environments that concurrently execute heterogeneous query workloads. At the same time, these systems need to satisfy diverse performance expectations. In these newly-emerging settings, avoiding potential Quality-of-Service (QoS) violations heavily relies on performance predictability, i.e., the ability to estimate the impact of concurrent query execution on the performance of individual queries in a continuously evolving workload. This paper presents a modeling approach to estimate the impact of concurrency on query performance for analytical workloads. Our solution relies on the analysis of query behavior in isolation, pairwise query interactions and sampling techniques to predict resource contention under various query mixes and concurrency levels. We introduce a simple yet powerful metric that accurately captures the joint effects of disk and memory contention on query performance in a single value. We also discuss predicting the execution behavior of a time-varying query workload through query-interaction timelines, i.e., a fine-grained estimation of the time segments during which discrete mixes will be executed concurrently. Our experimental evaluation on top of PostgreSQL/TPC-H demonstrates that our models can provide query latency predictions within approximately 20% of the actual values in the average case.

AB - Current trends in data management systems, such as cloud and multi-tenant databases, are leading to data processing environments that concurrently execute heterogeneous query workloads. At the same time, these systems need to satisfy diverse performance expectations. In these newly-emerging settings, avoiding potential Quality-of-Service (QoS) violations heavily relies on performance predictability, i.e., the ability to estimate the impact of concurrent query execution on the performance of individual queries in a continuously evolving workload. This paper presents a modeling approach to estimate the impact of concurrency on query performance for analytical workloads. Our solution relies on the analysis of query behavior in isolation, pairwise query interactions and sampling techniques to predict resource contention under various query mixes and concurrency levels. We introduce a simple yet powerful metric that accurately captures the joint effects of disk and memory contention on query performance in a single value. We also discuss predicting the execution behavior of a time-varying query workload through query-interaction timelines, i.e., a fine-grained estimation of the time segments during which discrete mixes will be executed concurrently. Our experimental evaluation on top of PostgreSQL/TPC-H demonstrates that our models can provide query latency predictions within approximately 20% of the actual values in the average case.

KW - concurrency

KW - query performance prediction

UR - http://www.scopus.com/inward/record.url?scp=79960005600&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960005600&partnerID=8YFLogxK

U2 - 10.1145/1989323.1989359

DO - 10.1145/1989323.1989359

M3 - Conference contribution

SN - 9781450306614

T3 - Proceedings of the ACM SIGMOD International Conference on Management of Data

SP - 337

EP - 348

BT - Proceedings of SIGMOD 2011 and PODS 2011

ER -

Duggan J, Cetintemel U, Papaemmanouil O, Upfal E. Performance prediction for concurrent database workloads. In Proceedings of SIGMOD 2011 and PODS 2011. 2011. p. 337-348. (Proceedings of the ACM SIGMOD International Conference on Management of Data). https://doi.org/10.1145/1989323.1989359