SlideShare a Scribd company logo
OLTP
Benchmark
B Y: S A I K AT G A R A I
outline
Introduction And Architecture
Workload
Performance vs Cost Comparison
Evolving Workload Mixtures
Conclusion
Introduction
What is OLTP?
OLTP is an operational system that supports transaction-oriented applications in a 3-tier
architecture. It administers the day to day transaction of an organization. OLTP is basically focused
on query processing, maintaining data integrity in multi-access environments as well as
effectiveness that is measured by the total number of transactions per second. The full form of
OLTP is Online Transaction Processing.
What is benchmarking?
A Benchmark is the act of running a computer program, a set of programs, or other operations, in
order to assess the relative performance of an object, normally by running a number of standard
tests and trials against it.
Benchmark Model
OLTPArchitecture :
Requirements of OLTP :
Scalability: the ability to drive the system under test at high transactional rates (i.e., test for max
throughput);
Fine-Grained Rate Control: the ability to control the rate of requests with great precision;
Mixed and Evolving Workloads: the ability to support mixed workloads, and to change the rate,
composition, and access distribution of the workloads dynamically over time;
Synthetic and Real Data & Workloads: support for both synthetic and real data sets and workloads
(and thus both programmatically-generated and trace-based executions);
Ease of Deployment/Portability: the ability to deploy experimental settings and run experiments
easily on a variety of DBMSs and DBaaSs using an integrated framework;
Extensibility: the ability to extend the set of workloads and systems supported with a minimal
engineering effort (e.g., centralizing SQL dialect localizations);
Lightweight, Fine-Grained Statistics Gathering: the ability to collect detailed statistics of both client-
side activity and server-side resource utilization with minimum impact on the overall performance;
Workloads
AuctionMark
Epinions
JPAB
ResourceStresser
SEATS
TATP
TPC-C
Twitter
Wikipedia
YCSB
Profile info for the benchmark workloads
Performance v/s Cost Comparison
The OLTP-Bench is used to measure the performance-vs-cost ratio of a single
DBaaS provider
Using the YCSB and Wikipedia benchmarks, we used OLTP-Bench to deploy
databases on five different instance sizes on Amazon’s RDS platform
Performance v/s Cost Comparison
Performance v/s Cost Comparison
We then ran each benchmark separately at its maximum speed for a total of 30
minutes.
 During that time, OLTP-Bench calculates the average maximum sustained
throughput and the 95th percentile latency from the middle 20 minutes of the
experiment’s run.
Performance v/s Cost Comparison
Performance v/s Cost Comparison
As for the price/performance ratio, the results for YCSB, the L instance yields
the best cost/performance ratio, with good overall throughput and low latency.
As for the price/performance ratio, the results for Wikipedia suggest that the
XL-HM instance is the best choice for this workload. Although the 2XL-HM and
4XL-HM instances provide better performance, the additional cost incurred is
more .
Evolving Workload Mixtures
OLTP-Bench’s ability to smoothly evolve the transaction mixture during an
experiment.
Choosing YCSB as our target workload since it is composed of a series of
simple transactions.
Using MySQL we run the system.
Representation using charts
Comparing DBaaS Providers
Network latency is an important factor for OLTP workloads
In this example all of the workers on virtual machines hosted
in the same datacenter region
The test is done test to ensure fairness between the
DBaaS providers
Cost V/s Performance
Conclusion
Benchmarking has always been central for high-performance and efficient data
management system design, deployment and maintenance.
The advent of cloud-based data management solutions with their highly shared
resources, non-user controlled system tuning, and novel pricing schemes further
exacerbate the need for careful benchmarking.
Thank You

More Related Content

Similar to OLTP benchmark by Saikat Garai, Presidency college, Bangalore.

2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...
2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...
2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...
IEEEFINALYEARSTUDENTPROJECT
 
2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...
2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...
2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...
IEEEFINALSEMSTUDENTSPROJECTS
 
IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...
IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...
IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...
IEEEGLOBALSOFTSTUDENTPROJECTS
 
Building a new CTL model checker using Web Services
Building a new CTL model checker using Web ServicesBuilding a new CTL model checker using Web Services
Building a new CTL model checker using Web Services
infopapers
 
CS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docx
CS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docxCS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docx
CS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docx
faithxdunce63732
 
% Razones para alta disponibilidad
% Razones para alta disponibilidad% Razones para alta disponibilidad
% Razones para alta disponibilidad
CPVEN
 
VMware Report Draft v2.1
VMware Report Draft v2.1VMware Report Draft v2.1
VMware Report Draft v2.1
John White
 
Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...
Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...
Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...
Principled Technologies
 
Iaetsd appliances of harmonizing model in cloud
Iaetsd appliances of harmonizing model in cloudIaetsd appliances of harmonizing model in cloud
Iaetsd appliances of harmonizing model in cloud
Iaetsd Iaetsd
 
Data warehousing testing strategies cognos
Data warehousing testing strategies cognosData warehousing testing strategies cognos
Data warehousing testing strategies cognos
Sandeep Mehta
 
SPROJReport (1)
SPROJReport (1)SPROJReport (1)
SPROJReport (1)
Hashaam Mahboob
 
Distributed Performance testing by funkload
Distributed Performance testing by funkloadDistributed Performance testing by funkload
Distributed Performance testing by funkload
Akhil Singh
 
Oracle databasecapacityanalysisusingstatisticalmethods
Oracle databasecapacityanalysisusingstatisticalmethodsOracle databasecapacityanalysisusingstatisticalmethods
Oracle databasecapacityanalysisusingstatisticalmethods
Ajith Narayanan
 
Traffic Simulator
Traffic SimulatorTraffic Simulator
Traffic Simulator
gystell
 
Workload-Aware Data Management in Shared-Nothing Distributed OLTP Databases
Workload-Aware Data Management in Shared-Nothing Distributed OLTP DatabasesWorkload-Aware Data Management in Shared-Nothing Distributed OLTP Databases
Workload-Aware Data Management in Shared-Nothing Distributed OLTP Databases
Joarder Kamal
 
B2 2005 introduction_load_testing_blackboard_primer_draft
B2 2005 introduction_load_testing_blackboard_primer_draftB2 2005 introduction_load_testing_blackboard_primer_draft
B2 2005 introduction_load_testing_blackboard_primer_draft
Steve Feldman
 
Anton Serputko Workshop “Тестування продуктивності”
Anton Serputko Workshop “Тестування продуктивності”Anton Serputko Workshop “Тестування продуктивності”
Anton Serputko Workshop “Тестування продуктивності”
Dakiry
 
A multi path routing algorithm for ip
A multi path routing algorithm for ipA multi path routing algorithm for ip
A multi path routing algorithm for ip
Alvianus Dengen
 
Performance testing in scope of migration to cloud by Serghei Radov
Performance testing in scope of migration to cloud by Serghei RadovPerformance testing in scope of migration to cloud by Serghei Radov
Performance testing in scope of migration to cloud by Serghei Radov
Valeriia Maliarenko
 
Mahout low-overhead datacenter traffic management using end-host-based ...
Mahout   low-overhead  datacenter  traffic management  using end-host-based  ...Mahout   low-overhead  datacenter  traffic management  using end-host-based  ...
Mahout low-overhead datacenter traffic management using end-host-based ...
João Gabriel Lima
 

Similar to OLTP benchmark by Saikat Garai, Presidency college, Bangalore. (20)

2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...
2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...
2014 IEEE JAVA NETWORKING COMPUTING PROJECT Cost effective resource allocatio...
 
2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...
2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...
2014 IEEE JAVA NETWORKING PROJECT Cost effective resource allocation of overl...
 
IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...
IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...
IEEE 2014 JAVA NETWORKING PROJECTS Cost effective resource allocation of over...
 
Building a new CTL model checker using Web Services
Building a new CTL model checker using Web ServicesBuilding a new CTL model checker using Web Services
Building a new CTL model checker using Web Services
 
CS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docx
CS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docxCS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docx
CS 301 Computer ArchitectureStudent # 1 EID 09Kingdom of .docx
 
% Razones para alta disponibilidad
% Razones para alta disponibilidad% Razones para alta disponibilidad
% Razones para alta disponibilidad
 
VMware Report Draft v2.1
VMware Report Draft v2.1VMware Report Draft v2.1
VMware Report Draft v2.1
 
Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...
Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...
Three Microsoft Azure SQL Managed Instances offered better SQL Server perform...
 
Iaetsd appliances of harmonizing model in cloud
Iaetsd appliances of harmonizing model in cloudIaetsd appliances of harmonizing model in cloud
Iaetsd appliances of harmonizing model in cloud
 
Data warehousing testing strategies cognos
Data warehousing testing strategies cognosData warehousing testing strategies cognos
Data warehousing testing strategies cognos
 
SPROJReport (1)
SPROJReport (1)SPROJReport (1)
SPROJReport (1)
 
Distributed Performance testing by funkload
Distributed Performance testing by funkloadDistributed Performance testing by funkload
Distributed Performance testing by funkload
 
Oracle databasecapacityanalysisusingstatisticalmethods
Oracle databasecapacityanalysisusingstatisticalmethodsOracle databasecapacityanalysisusingstatisticalmethods
Oracle databasecapacityanalysisusingstatisticalmethods
 
Traffic Simulator
Traffic SimulatorTraffic Simulator
Traffic Simulator
 
Workload-Aware Data Management in Shared-Nothing Distributed OLTP Databases
Workload-Aware Data Management in Shared-Nothing Distributed OLTP DatabasesWorkload-Aware Data Management in Shared-Nothing Distributed OLTP Databases
Workload-Aware Data Management in Shared-Nothing Distributed OLTP Databases
 
B2 2005 introduction_load_testing_blackboard_primer_draft
B2 2005 introduction_load_testing_blackboard_primer_draftB2 2005 introduction_load_testing_blackboard_primer_draft
B2 2005 introduction_load_testing_blackboard_primer_draft
 
Anton Serputko Workshop “Тестування продуктивності”
Anton Serputko Workshop “Тестування продуктивності”Anton Serputko Workshop “Тестування продуктивності”
Anton Serputko Workshop “Тестування продуктивності”
 
A multi path routing algorithm for ip
A multi path routing algorithm for ipA multi path routing algorithm for ip
A multi path routing algorithm for ip
 
Performance testing in scope of migration to cloud by Serghei Radov
Performance testing in scope of migration to cloud by Serghei RadovPerformance testing in scope of migration to cloud by Serghei Radov
Performance testing in scope of migration to cloud by Serghei Radov
 
Mahout low-overhead datacenter traffic management using end-host-based ...
Mahout   low-overhead  datacenter  traffic management  using end-host-based  ...Mahout   low-overhead  datacenter  traffic management  using end-host-based  ...
Mahout low-overhead datacenter traffic management using end-host-based ...
 

Recently uploaded

“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
Edge AI and Vision Alliance
 
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-EfficiencyFreshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
ScyllaDB
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
Tatiana Kojar
 
9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...
9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...
9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...
saastr
 
Essentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation ParametersEssentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation Parameters
Safe Software
 
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansBiomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Neo4j
 
Apps Break Data
Apps Break DataApps Break Data
Apps Break Data
Ivo Velitchkov
 
GraphRAG for LifeSciences Hands-On with the Clinical Knowledge Graph
GraphRAG for LifeSciences Hands-On with the Clinical Knowledge GraphGraphRAG for LifeSciences Hands-On with the Clinical Knowledge Graph
GraphRAG for LifeSciences Hands-On with the Clinical Knowledge Graph
Neo4j
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
Miro Wengner
 
Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
Jason Packer
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
AstuteBusiness
 
Principle of conventional tomography-Bibash Shahi ppt..pptx
Principle of conventional tomography-Bibash Shahi ppt..pptxPrinciple of conventional tomography-Bibash Shahi ppt..pptx
Principle of conventional tomography-Bibash Shahi ppt..pptx
BibashShahi
 
Dandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity serverDandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity server
Antonios Katsarakis
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
Jakub Marek
 
zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...
zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...
zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...
Alex Pruden
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
DianaGray10
 
The Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptxThe Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptx
operationspcvita
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
Zilliz
 
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdfMonitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Tosin Akinosho
 

Recently uploaded (20)

“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
 
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-EfficiencyFreshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
 
9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...
9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...
9 CEO's who hit $100m ARR Share Their Top Growth Tactics Nathan Latka, Founde...
 
Essentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation ParametersEssentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation Parameters
 
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansBiomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
 
Apps Break Data
Apps Break DataApps Break Data
Apps Break Data
 
Artificial Intelligence and Electronic Warfare
Artificial Intelligence and Electronic WarfareArtificial Intelligence and Electronic Warfare
Artificial Intelligence and Electronic Warfare
 
GraphRAG for LifeSciences Hands-On with the Clinical Knowledge Graph
GraphRAG for LifeSciences Hands-On with the Clinical Knowledge GraphGraphRAG for LifeSciences Hands-On with the Clinical Knowledge Graph
GraphRAG for LifeSciences Hands-On with the Clinical Knowledge Graph
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
 
Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
 
Principle of conventional tomography-Bibash Shahi ppt..pptx
Principle of conventional tomography-Bibash Shahi ppt..pptxPrinciple of conventional tomography-Bibash Shahi ppt..pptx
Principle of conventional tomography-Bibash Shahi ppt..pptx
 
Dandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity serverDandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity server
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
 
zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...
zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...
zkStudyClub - LatticeFold: A Lattice-based Folding Scheme and its Application...
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
 
The Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptxThe Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptx
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
 
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdfMonitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdf
 

OLTP benchmark by Saikat Garai, Presidency college, Bangalore.

  • 1. OLTP Benchmark B Y: S A I K AT G A R A I
  • 2. outline Introduction And Architecture Workload Performance vs Cost Comparison Evolving Workload Mixtures Conclusion
  • 3. Introduction What is OLTP? OLTP is an operational system that supports transaction-oriented applications in a 3-tier architecture. It administers the day to day transaction of an organization. OLTP is basically focused on query processing, maintaining data integrity in multi-access environments as well as effectiveness that is measured by the total number of transactions per second. The full form of OLTP is Online Transaction Processing. What is benchmarking? A Benchmark is the act of running a computer program, a set of programs, or other operations, in order to assess the relative performance of an object, normally by running a number of standard tests and trials against it.
  • 7. Scalability: the ability to drive the system under test at high transactional rates (i.e., test for max throughput); Fine-Grained Rate Control: the ability to control the rate of requests with great precision; Mixed and Evolving Workloads: the ability to support mixed workloads, and to change the rate, composition, and access distribution of the workloads dynamically over time; Synthetic and Real Data & Workloads: support for both synthetic and real data sets and workloads (and thus both programmatically-generated and trace-based executions); Ease of Deployment/Portability: the ability to deploy experimental settings and run experiments easily on a variety of DBMSs and DBaaSs using an integrated framework; Extensibility: the ability to extend the set of workloads and systems supported with a minimal engineering effort (e.g., centralizing SQL dialect localizations); Lightweight, Fine-Grained Statistics Gathering: the ability to collect detailed statistics of both client- side activity and server-side resource utilization with minimum impact on the overall performance;
  • 9. Profile info for the benchmark workloads
  • 10. Performance v/s Cost Comparison The OLTP-Bench is used to measure the performance-vs-cost ratio of a single DBaaS provider Using the YCSB and Wikipedia benchmarks, we used OLTP-Bench to deploy databases on five different instance sizes on Amazon’s RDS platform
  • 11. Performance v/s Cost Comparison
  • 12. Performance v/s Cost Comparison We then ran each benchmark separately at its maximum speed for a total of 30 minutes.  During that time, OLTP-Bench calculates the average maximum sustained throughput and the 95th percentile latency from the middle 20 minutes of the experiment’s run.
  • 13. Performance v/s Cost Comparison
  • 14. Performance v/s Cost Comparison As for the price/performance ratio, the results for YCSB, the L instance yields the best cost/performance ratio, with good overall throughput and low latency. As for the price/performance ratio, the results for Wikipedia suggest that the XL-HM instance is the best choice for this workload. Although the 2XL-HM and 4XL-HM instances provide better performance, the additional cost incurred is more .
  • 15. Evolving Workload Mixtures OLTP-Bench’s ability to smoothly evolve the transaction mixture during an experiment. Choosing YCSB as our target workload since it is composed of a series of simple transactions. Using MySQL we run the system.
  • 17. Comparing DBaaS Providers Network latency is an important factor for OLTP workloads In this example all of the workers on virtual machines hosted in the same datacenter region The test is done test to ensure fairness between the DBaaS providers
  • 19. Conclusion Benchmarking has always been central for high-performance and efficient data management system design, deployment and maintenance. The advent of cloud-based data management solutions with their highly shared resources, non-user controlled system tuning, and novel pricing schemes further exacerbate the need for careful benchmarking.