SlideShare une entreprise Scribd logo
1  sur  15
Télécharger pour lire hors ligne
Economic Scheduling of
Hadoop Jobs
-The Dynamic Priority MapReduce Scheduler



Thomas Sandholm, HP Labs, Palo Alto
Kevin Lai, HP Labs, Palo Alto




                                            1
The Problem

»   Allocate slots on compute nodes for job tasks
»   Classic Approach: Throughput optimization
»   Cross User Priorities inferred based on heuristics
»   Social Scheduling
»   Our Approach: User value optimization
»   Users are given an incentive to scale up or down
»   Automate demand conflict resolution




                                                         2
Other Hadoop Schedulers

»   FIFO
»   HOD
»   Fairshare
»   Capacity

» Designed for no queues or few static fixed QoS
  queues
» Works well in corporate clusters




                                                   3
Dynamic Priority Scheduler Requirements

» Users may come and go frequently
» Users may be unknown to providers
» Users may want to schedule jobs across data
  centers and Hadoop installations




         manual, social scheduling of users
     (assumed to be cooperating) breaks down



                                                4
Architecture




               4/(4+1.5+2)*15=8




                                  5
Our Solution: Automated Resource Allocation




                               Budget Remaining




Running                              Share
Tasks

 Pending                  Spending Rate
 Tasks
                                                  6
Proportional-Share Scheduling

» qi = bi/(bi + p)
» p = ∑ b-i

» Huberman et al Spawn ‘92
» Waldspurger et al Lottery Scheduling ‘95
» Lai et al Tycoon ‘05




                                             7
Key Design Principles

» Pay-per-use: spending rate is only deducted from budget
  if a job performed work
» Work-conserving: users are never charged more than
  their spending rates but can get more slots if other users
  are idle
» Preemptive: higher spending users may cause tasks from
  lower spending users to be killed
» Scalable: No memory, or history-based fair-share
  smoothing




                                                        8
Implementation

» Standalone Hadoop MapReduce JobTracker Scheduler
  Plugin
» HTTP/XML/REST Servlet to provide secure management
  and monitoring of queues
» Generic queue allocation/accounting classes (could
  move into mapred core)
» Pluggable scheduler enforcing shares, when scheduling
  jobs (could be replaced by capacity/fairshare
  enforcers)




                                                    9
Configuration
Option                                    Examples
mapred.jobtracker.taskScheduler           org.apache.hadoop.mapred.DynamicPriorityScheduler



mapred.priority-scheduler.kill-interval   0



mapred.dynamic-scheduler.alloc-interval   20



mapred.dynamic-scheduler.budget-file      /etc/hadoop.budget



mapred.priority-scheduler.acl-file        /etc/hadoop.acl




                                                                                      10
Experiment


 Fairshare vs Capacity vs FIFO vs DP
 2-80 simulated users/queues
 2 Clusters
 PiEstimator Simulation




                                       11
Budget Dynamics
                                        DynPrio preempt FIFO scheduler
                     DynPrio no                   Capacity scheduler
                     preempt




Funding runs out   Budget replenished


                                                                         12
Service Differentiation




               DynPrio    FIFO

                                 13
Dynamic Adjustment




                     14
More info

» Papers
  › SIGMETRICS 2009
  › Workshop on Job Scheduling for Parallel Processing
    (JSSPP’10)
  › International Conference on Cloud Computing and
    Virtualization (CCV’10)
» HADOOP-4768 JIRA
» Source:
  http://svn.apache.org/viewvc/hadoop/mapreduce/trunk/src/contrib/dynamic-scheduler/




                                                                                  15

Contenu connexe

Tendances

Tendances (20)

Self adjusting slot configurations for homogeneous and heterogeneous hadoop c...
Self adjusting slot configurations for homogeneous and heterogeneous hadoop c...Self adjusting slot configurations for homogeneous and heterogeneous hadoop c...
Self adjusting slot configurations for homogeneous and heterogeneous hadoop c...
 
Hadoop Summit San Jose 2014: Costing Your Big Data Operations
Hadoop Summit San Jose 2014: Costing Your Big Data Operations Hadoop Summit San Jose 2014: Costing Your Big Data Operations
Hadoop Summit San Jose 2014: Costing Your Big Data Operations
 
Advanced Hadoop Tuning and Optimization
Advanced Hadoop Tuning and Optimization Advanced Hadoop Tuning and Optimization
Advanced Hadoop Tuning and Optimization
 
Hadoop scheduler
Hadoop schedulerHadoop scheduler
Hadoop scheduler
 
Distributed Processing Frameworks
Distributed Processing FrameworksDistributed Processing Frameworks
Distributed Processing Frameworks
 
Analysing of big data using map reduce
Analysing of big data using map reduceAnalysing of big data using map reduce
Analysing of big data using map reduce
 
Resource Aware Scheduling for Hadoop [Final Presentation]
Resource Aware Scheduling for Hadoop [Final Presentation]Resource Aware Scheduling for Hadoop [Final Presentation]
Resource Aware Scheduling for Hadoop [Final Presentation]
 
Capacity Management and BigData/Hadoop - Hitchhiker's guide for the Capacity ...
Capacity Management and BigData/Hadoop - Hitchhiker's guide for the Capacity ...Capacity Management and BigData/Hadoop - Hitchhiker's guide for the Capacity ...
Capacity Management and BigData/Hadoop - Hitchhiker's guide for the Capacity ...
 
MapReduce Scheduling Algorithms
MapReduce Scheduling AlgorithmsMapReduce Scheduling Algorithms
MapReduce Scheduling Algorithms
 
Anju
AnjuAnju
Anju
 
Present & Future of Greenplum Database A massively parallel Postgres Database...
Present & Future of Greenplum Database A massively parallel Postgres Database...Present & Future of Greenplum Database A massively parallel Postgres Database...
Present & Future of Greenplum Database A massively parallel Postgres Database...
 
Greenplum-Spark November 2018
Greenplum-Spark November 2018Greenplum-Spark November 2018
Greenplum-Spark November 2018
 
May 2013 HUG: HCatalog/Hive Data Out
May 2013 HUG: HCatalog/Hive Data OutMay 2013 HUG: HCatalog/Hive Data Out
May 2013 HUG: HCatalog/Hive Data Out
 
Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...
Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...
Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...
 
Hadoop Summit Brussels 2015: Architecting a Scalable Hadoop Platform - Top 10...
Hadoop Summit Brussels 2015: Architecting a Scalable Hadoop Platform - Top 10...Hadoop Summit Brussels 2015: Architecting a Scalable Hadoop Platform - Top 10...
Hadoop Summit Brussels 2015: Architecting a Scalable Hadoop Platform - Top 10...
 
Machine Learning, Graph, Text and Geospatial on Postgres and Greenplum - Gree...
Machine Learning, Graph, Text and Geospatial on Postgres and Greenplum - Gree...Machine Learning, Graph, Text and Geospatial on Postgres and Greenplum - Gree...
Machine Learning, Graph, Text and Geospatial on Postgres and Greenplum - Gree...
 
Introduction to Hadoop and MapReduce
Introduction to Hadoop and MapReduceIntroduction to Hadoop and MapReduce
Introduction to Hadoop and MapReduce
 
Tensor Processing Unit (TPU)
Tensor Processing Unit (TPU)Tensor Processing Unit (TPU)
Tensor Processing Unit (TPU)
 
Mixing Analytic Workloads with Greenplum and Apache Spark
Mixing Analytic Workloads with Greenplum and Apache SparkMixing Analytic Workloads with Greenplum and Apache Spark
Mixing Analytic Workloads with Greenplum and Apache Spark
 
Greenplum for Kubernetes - Greenplum Summit 2019
Greenplum for Kubernetes - Greenplum Summit 2019Greenplum for Kubernetes - Greenplum Summit 2019
Greenplum for Kubernetes - Greenplum Summit 2019
 

En vedette

Battle At Goliad
Battle At GoliadBattle At Goliad
Battle At Goliad
compd
 
2013 11-19-hoya-status
2013 11-19-hoya-status2013 11-19-hoya-status
2013 11-19-hoya-status
Steve Loughran
 
Taming Deployment With Smart Frog
Taming Deployment With Smart FrogTaming Deployment With Smart Frog
Taming Deployment With Smart Frog
Steve Loughran
 
A New Approach To Organization
A New Approach To OrganizationA New Approach To Organization
A New Approach To Organization
compd
 

En vedette (14)

My other computer_is_a_datacentre
My other computer_is_a_datacentreMy other computer_is_a_datacentre
My other computer_is_a_datacentre
 
Extended essay overview
Extended essay overviewExtended essay overview
Extended essay overview
 
Battle At Goliad
Battle At GoliadBattle At Goliad
Battle At Goliad
 
H is for_hadoop
H is for_hadoopH is for_hadoop
H is for_hadoop
 
Overview of slider project
Overview of slider projectOverview of slider project
Overview of slider project
 
Farms, Fabrics and Clouds
Farms, Fabrics and CloudsFarms, Fabrics and Clouds
Farms, Fabrics and Clouds
 
2013 11-19-hoya-status
2013 11-19-hoya-status2013 11-19-hoya-status
2013 11-19-hoya-status
 
Taming Deployment With Smart Frog
Taming Deployment With Smart FrogTaming Deployment With Smart Frog
Taming Deployment With Smart Frog
 
HDP-1 introduction for HUG France
HDP-1 introduction for HUG FranceHDP-1 introduction for HUG France
HDP-1 introduction for HUG France
 
A New Approach To Organization
A New Approach To OrganizationA New Approach To Organization
A New Approach To Organization
 
Graphs
GraphsGraphs
Graphs
 
Scholarly articles
Scholarly articlesScholarly articles
Scholarly articles
 
Echolocation
EcholocationEcholocation
Echolocation
 
Did you really want that data?
Did you really want that data?Did you really want that data?
Did you really want that data?
 

Similaire à Economic Scheduling of Hadoop Jobs

SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013
SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013
SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013
James McGalliard
 
Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015
Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015
Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015
Deanna Kosaraju
 
Introduction to Apache Hadoop
Introduction to Apache HadoopIntroduction to Apache Hadoop
Introduction to Apache Hadoop
Christopher Pezza
 
Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...
Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...
Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...
DataWorks Summit
 

Similaire à Economic Scheduling of Hadoop Jobs (20)

Apache Hadoop YARN - The Future of Data Processing with Hadoop
Apache Hadoop YARN - The Future of Data Processing with HadoopApache Hadoop YARN - The Future of Data Processing with Hadoop
Apache Hadoop YARN - The Future of Data Processing with Hadoop
 
SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013
SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013
SOME WORKLOAD SCHEDULING ALTERNATIVES 11.07.2013
 
Apache Airflow (incubating) NL HUG Meetup 2016-07-19
Apache Airflow (incubating) NL HUG Meetup 2016-07-19Apache Airflow (incubating) NL HUG Meetup 2016-07-19
Apache Airflow (incubating) NL HUG Meetup 2016-07-19
 
L017656475
L017656475L017656475
L017656475
 
Map-Reduce Synchronized and Comparative Queue Capacity Scheduler in Hadoop fo...
Map-Reduce Synchronized and Comparative Queue Capacity Scheduler in Hadoop fo...Map-Reduce Synchronized and Comparative Queue Capacity Scheduler in Hadoop fo...
Map-Reduce Synchronized and Comparative Queue Capacity Scheduler in Hadoop fo...
 
Oct 2011 CHADNUG Presentation on Hadoop
Oct 2011 CHADNUG Presentation on HadoopOct 2011 CHADNUG Presentation on Hadoop
Oct 2011 CHADNUG Presentation on Hadoop
 
Dache: A Data Aware Caching for Big-Data Applications Using the MapReduce Fra...
Dache: A Data Aware Caching for Big-Data Applications Usingthe MapReduce Fra...Dache: A Data Aware Caching for Big-Data Applications Usingthe MapReduce Fra...
Dache: A Data Aware Caching for Big-Data Applications Using the MapReduce Fra...
 
Review of Calculation Paradigm and its Components
Review of Calculation Paradigm and its ComponentsReview of Calculation Paradigm and its Components
Review of Calculation Paradigm and its Components
 
Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015
Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015
Optimal Execution Of MapReduce Jobs In Cloud - Voices 2015
 
Hadoop Summit San Jose 2015: What it Takes to Run Hadoop at Scale Yahoo Persp...
Hadoop Summit San Jose 2015: What it Takes to Run Hadoop at Scale Yahoo Persp...Hadoop Summit San Jose 2015: What it Takes to Run Hadoop at Scale Yahoo Persp...
Hadoop Summit San Jose 2015: What it Takes to Run Hadoop at Scale Yahoo Persp...
 
Introduction to Hadoop
Introduction to HadoopIntroduction to Hadoop
Introduction to Hadoop
 
Hadoop tutorial
Hadoop tutorialHadoop tutorial
Hadoop tutorial
 
Apache Tez : Accelerating Hadoop Query Processing
Apache Tez : Accelerating Hadoop Query ProcessingApache Tez : Accelerating Hadoop Query Processing
Apache Tez : Accelerating Hadoop Query Processing
 
Hadoop Tutorial.ppt
Hadoop Tutorial.pptHadoop Tutorial.ppt
Hadoop Tutorial.ppt
 
Hadoop ensma poitiers
Hadoop ensma poitiersHadoop ensma poitiers
Hadoop ensma poitiers
 
Introduction to Apache Hadoop
Introduction to Apache HadoopIntroduction to Apache Hadoop
Introduction to Apache Hadoop
 
Track B-3 解構大數據架構 - 大數據系統的伺服器與網路資源規劃
Track B-3 解構大數據架構 - 大數據系統的伺服器與網路資源規劃Track B-3 解構大數據架構 - 大數據系統的伺服器與網路資源規劃
Track B-3 解構大數據架構 - 大數據系統的伺服器與網路資源規劃
 
Bigdata and Hadoop with Docker
Bigdata and Hadoop with DockerBigdata and Hadoop with Docker
Bigdata and Hadoop with Docker
 
Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...
Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...
Dr Elephant: LinkedIn's Self-Service System for Detecting and Treating Hadoop...
 
Drill into Drill – How Providing Flexibility and Performance is Possible
Drill into Drill – How Providing Flexibility and Performance is PossibleDrill into Drill – How Providing Flexibility and Performance is Possible
Drill into Drill – How Providing Flexibility and Performance is Possible
 

Plus de Steve Loughran

Plus de Steve Loughran (20)

Hadoop Vectored IO
Hadoop Vectored IOHadoop Vectored IO
Hadoop Vectored IO
 
The age of rename() is over
The age of rename() is overThe age of rename() is over
The age of rename() is over
 
What does Rename Do: (detailed version)
What does Rename Do: (detailed version)What does Rename Do: (detailed version)
What does Rename Do: (detailed version)
 
Put is the new rename: San Jose Summit Edition
Put is the new rename: San Jose Summit EditionPut is the new rename: San Jose Summit Edition
Put is the new rename: San Jose Summit Edition
 
@Dissidentbot: dissent will be automated!
@Dissidentbot: dissent will be automated!@Dissidentbot: dissent will be automated!
@Dissidentbot: dissent will be automated!
 
PUT is the new rename()
PUT is the new rename()PUT is the new rename()
PUT is the new rename()
 
Extreme Programming Deployed
Extreme Programming DeployedExtreme Programming Deployed
Extreme Programming Deployed
 
Testing
TestingTesting
Testing
 
I hate mocking
I hate mockingI hate mocking
I hate mocking
 
What does rename() do?
What does rename() do?What does rename() do?
What does rename() do?
 
Dancing Elephants: Working with Object Storage in Apache Spark and Hive
Dancing Elephants: Working with Object Storage in Apache Spark and HiveDancing Elephants: Working with Object Storage in Apache Spark and Hive
Dancing Elephants: Working with Object Storage in Apache Spark and Hive
 
Apache Spark and Object Stores —for London Spark User Group
Apache Spark and Object Stores —for London Spark User GroupApache Spark and Object Stores —for London Spark User Group
Apache Spark and Object Stores —for London Spark User Group
 
Spark Summit East 2017: Apache spark and object stores
Spark Summit East 2017: Apache spark and object storesSpark Summit East 2017: Apache spark and object stores
Spark Summit East 2017: Apache spark and object stores
 
Hadoop, Hive, Spark and Object Stores
Hadoop, Hive, Spark and Object StoresHadoop, Hive, Spark and Object Stores
Hadoop, Hive, Spark and Object Stores
 
Apache Spark and Object Stores
Apache Spark and Object StoresApache Spark and Object Stores
Apache Spark and Object Stores
 
Household INFOSEC in a Post-Sony Era
Household INFOSEC in a Post-Sony EraHousehold INFOSEC in a Post-Sony Era
Household INFOSEC in a Post-Sony Era
 
Hadoop and Kerberos: the Madness Beyond the Gate: January 2016 edition
Hadoop and Kerberos: the Madness Beyond the Gate: January 2016 editionHadoop and Kerberos: the Madness Beyond the Gate: January 2016 edition
Hadoop and Kerberos: the Madness Beyond the Gate: January 2016 edition
 
Hadoop and Kerberos: the Madness Beyond the Gate
Hadoop and Kerberos: the Madness Beyond the GateHadoop and Kerberos: the Madness Beyond the Gate
Hadoop and Kerberos: the Madness Beyond the Gate
 
Slider: Applications on YARN
Slider: Applications on YARNSlider: Applications on YARN
Slider: Applications on YARN
 
YARN Services
YARN ServicesYARN Services
YARN Services
 

Dernier

Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Victor Rentea
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 

Dernier (20)

Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectors
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Cyberprint. Dark Pink Apt Group [EN].pdf
Cyberprint. Dark Pink Apt Group [EN].pdfCyberprint. Dark Pink Apt Group [EN].pdf
Cyberprint. Dark Pink Apt Group [EN].pdf
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
 

Economic Scheduling of Hadoop Jobs

  • 1. Economic Scheduling of Hadoop Jobs -The Dynamic Priority MapReduce Scheduler Thomas Sandholm, HP Labs, Palo Alto Kevin Lai, HP Labs, Palo Alto 1
  • 2. The Problem » Allocate slots on compute nodes for job tasks » Classic Approach: Throughput optimization » Cross User Priorities inferred based on heuristics » Social Scheduling » Our Approach: User value optimization » Users are given an incentive to scale up or down » Automate demand conflict resolution 2
  • 3. Other Hadoop Schedulers » FIFO » HOD » Fairshare » Capacity » Designed for no queues or few static fixed QoS queues » Works well in corporate clusters 3
  • 4. Dynamic Priority Scheduler Requirements » Users may come and go frequently » Users may be unknown to providers » Users may want to schedule jobs across data centers and Hadoop installations manual, social scheduling of users (assumed to be cooperating) breaks down 4
  • 5. Architecture 4/(4+1.5+2)*15=8 5
  • 6. Our Solution: Automated Resource Allocation Budget Remaining Running Share Tasks Pending Spending Rate Tasks 6
  • 7. Proportional-Share Scheduling » qi = bi/(bi + p) » p = ∑ b-i » Huberman et al Spawn ‘92 » Waldspurger et al Lottery Scheduling ‘95 » Lai et al Tycoon ‘05 7
  • 8. Key Design Principles » Pay-per-use: spending rate is only deducted from budget if a job performed work » Work-conserving: users are never charged more than their spending rates but can get more slots if other users are idle » Preemptive: higher spending users may cause tasks from lower spending users to be killed » Scalable: No memory, or history-based fair-share smoothing 8
  • 9. Implementation » Standalone Hadoop MapReduce JobTracker Scheduler Plugin » HTTP/XML/REST Servlet to provide secure management and monitoring of queues » Generic queue allocation/accounting classes (could move into mapred core) » Pluggable scheduler enforcing shares, when scheduling jobs (could be replaced by capacity/fairshare enforcers) 9
  • 10. Configuration Option Examples mapred.jobtracker.taskScheduler org.apache.hadoop.mapred.DynamicPriorityScheduler mapred.priority-scheduler.kill-interval 0 mapred.dynamic-scheduler.alloc-interval 20 mapred.dynamic-scheduler.budget-file /etc/hadoop.budget mapred.priority-scheduler.acl-file /etc/hadoop.acl 10
  • 11. Experiment Fairshare vs Capacity vs FIFO vs DP 2-80 simulated users/queues 2 Clusters PiEstimator Simulation 11
  • 12. Budget Dynamics DynPrio preempt FIFO scheduler DynPrio no Capacity scheduler preempt Funding runs out Budget replenished 12
  • 13. Service Differentiation DynPrio FIFO 13
  • 15. More info » Papers › SIGMETRICS 2009 › Workshop on Job Scheduling for Parallel Processing (JSSPP’10) › International Conference on Cloud Computing and Virtualization (CCV’10) » HADOOP-4768 JIRA » Source: http://svn.apache.org/viewvc/hadoop/mapreduce/trunk/src/contrib/dynamic-scheduler/ 15