HUG Nov 2010: HDFS Raid - Facebook

Yahoo Developer Network
Yahoo Developer NetworkYahoo Developer Network
HDFS RAID,[object Object],DhrubaBorthakur (dhruba@fb.com),[object Object],Rodrigo Schmidt (rschmidt@fb.com),[object Object],RamkumarVadali (rvadali@fb.com),[object Object],Scott Chen (schen@fb.com),[object Object],Patrick  Kling (pkling@fb.com),[object Object]
Agenda,[object Object],What is RAID,[object Object],RAID at Facebook,[object Object],Anatomy of RAID,[object Object],How to Deploy,[object Object],Questions,[object Object]
What Is RAID,[object Object],[object Object],Default HDFS replication is 3,[object Object],Too much at PetaByte scale,[object Object],RAID helps save space in HDFS,[object Object],Reduce replication of “source” data,[object Object],Data safety using “parity” data,[object Object]
Tolerates 2 missing blocks, Storage cost 3x ,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],Tolerates 4 missing blocks, Storage cost 1.4x ,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],Source file,[object Object],P1,[object Object],P2,[object Object],P3,[object Object],P4,[object Object],Parity file,[object Object],Reed-Solomon Erasure Codes,[object Object]
RAID at Facebook,[object Object],Reduces disk usage in the warehouse,[object Object],Currently saving about 5PB with XOR RAID,[object Object],Gradual deployment,[object Object],Started with few tables,[object Object],Now used with all tables,[object Object],Reed Solomon RAID under way,[object Object]
Saving 5PB at Facebook,[object Object]
Anatomy of RAID,[object Object],Server-side:,[object Object],RaidNode,[object Object],BlockFixer,[object Object],Block placement policy,[object Object],Client-side:,[object Object],DistributedRaidFileSystem,[object Object],Raid Shell,[object Object]
Anatomy of RAID,[object Object],DataNodes,[object Object],NameNode,[object Object],[object Object]
 Get files to raidRaidNode,[object Object],[object Object]
 Fix missing blocks
 Recover files while readingJobTracker,[object Object],Raid File System,[object Object]
RaidNode,[object Object],Daemon that scans filesystem,[object Object],Policy file used to provide file patterns,[object Object],Generate parity files,[object Object],Single thread,[object Object],Map-Reduce job,[object Object],Reduces replication of source file,[object Object],One thread to purge outdated parity files,[object Object],If the source gets deleted,[object Object],One thread to HAR parity files,[object Object],To reduce inode count,[object Object]
Block Fixer,[object Object],Reconstructs missing/corrupt blocks,[object Object],Retrieves a list of corrupt files from NameNode,[object Object],Source blocks are reconstructed by “decoding”,[object Object],Parity blocks are reconstructed by “encoding”,[object Object]
Block Fixer,[object Object],Bonus: Parity HARs,[object Object],One HAR block => multiple parity blocks,[object Object],Reconstructs all necessary blocks,[object Object]
Block Fixer Stats,[object Object]
Erasure Code,[object Object],ErasureCode,[object Object],abstraction for erasure code implementations,[object Object],   public void encode(int[] message, int[] parity);,[object Object],   public void decode(int[] data, ,[object Object],int[] erasedLocations,,[object Object],int[] erasedValues);,[object Object],Current implementations,[object Object],XOR Code,[object Object],Reed Solomon Code,[object Object],Encoder/Decoder – uses ErasureCode to integrate with RAID framework,[object Object]
Block Placement,[object Object],Replication = 3, Tolerates any 2 errors,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],Dependent Blocks,[object Object],Replication = 1, Parity Length = 4, Tolerates any 4 errors,[object Object],1,[object Object],2,[object Object],3,[object Object],4,[object Object],5,[object Object],6,[object Object],7,[object Object],8,[object Object],9,[object Object],10,[object Object],P1,[object Object],P2,[object Object],P3,[object Object],P4,[object Object],Dependent Blocks,[object Object]
Block Placement,[object Object],Raid introduces new dependency between blocks in source and parity files,[object Object],Default block placement is bad for RAID,[object Object],Source/Parity blocks can be on a single node/rack,[object Object],Parity blocks could co-locate with source blocks,[object Object],Raid Block Policy,[object Object],Source files: After RAIDing, disperse blocks,[object Object],Parity files: Control placement of parity blocks to avoid source blocks and other parity blocks,[object Object]
DistributedRaidFileSystem,[object Object],A filter file system implementation,[object Object],Allows clients to read “corrupt” source files,[object Object],Catches BlockMissingException, ChecksumException,[object Object],Recreates missing blocks on the fly by using parity,[object Object],Does not fix the missing blocks,[object Object],Only allows the reads to succeed,[object Object]
RaidShell,[object Object],Administrator tool,[object Object],Recover blocks,[object Object],Reconstruct missing blocks,[object Object],Send reconstructed block to a data node,[object Object],Raid FSCK,[object Object], Report corrupt files that cannot be fixed by raid,[object Object],Handy tool as a last resort to fix blocks,[object Object]
Deployment,[object Object],Single configuration file “raid.xml”,[object Object],Specifies file patterns to RAID,[object Object],In HDFS config file,[object Object],Specify raid.xml location,[object Object],Specify location of parity files (default: /raid),[object Object],Specify FileSystem, BlockPlacementPolicy,[object Object],Starting RaidNode,[object Object],start-raidnode.sh, stop-raidnode.sh,[object Object],http://wiki.apache.org/hadoop/HDFS-RAID,[object Object]
1 of 21

Recommended

Nov 2010 HUG: Fuzzy Table - B.A.H by
Nov 2010 HUG: Fuzzy Table - B.A.HNov 2010 HUG: Fuzzy Table - B.A.H
Nov 2010 HUG: Fuzzy Table - B.A.HYahoo Developer Network
2.9K views47 slides
File Context by
File ContextFile Context
File ContextHadoop User Group
2.5K views17 slides
DataEngConf: Uri Laserson (Data Scientist, Cloudera) Scaling up Genomics with... by
DataEngConf: Uri Laserson (Data Scientist, Cloudera) Scaling up Genomics with...DataEngConf: Uri Laserson (Data Scientist, Cloudera) Scaling up Genomics with...
DataEngConf: Uri Laserson (Data Scientist, Cloudera) Scaling up Genomics with...Hakka Labs
512 views52 slides
January 2011 HUG: Howl Presentation by
January 2011 HUG: Howl PresentationJanuary 2011 HUG: Howl Presentation
January 2011 HUG: Howl PresentationYahoo Developer Network
5.3K views20 slides
January 2011 HUG: Pig Presentation by
January 2011 HUG: Pig PresentationJanuary 2011 HUG: Pig Presentation
January 2011 HUG: Pig PresentationYahoo Developer Network
3.4K views15 slides
Hadoop Adventures At Spotify (Strata Conference + Hadoop World 2013) by
Hadoop Adventures At Spotify (Strata Conference + Hadoop World 2013)Hadoop Adventures At Spotify (Strata Conference + Hadoop World 2013)
Hadoop Adventures At Spotify (Strata Conference + Hadoop World 2013)Adam Kawa
41.5K views82 slides

More Related Content

What's hot

Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T... by
Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T...Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T...
Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T...Spark Summit
2K views34 slides
Burst Presto & Spark workloads to AWS EMR with no data copies by
Burst Presto & Spark workloads to AWS EMR with no data copiesBurst Presto & Spark workloads to AWS EMR with no data copies
Burst Presto & Spark workloads to AWS EMR with no data copiesAlluxio, Inc.
1.8K views14 slides
CaffeOnSpark Update: Recent Enhancements and Use Cases by
CaffeOnSpark Update: Recent Enhancements and Use CasesCaffeOnSpark Update: Recent Enhancements and Use Cases
CaffeOnSpark Update: Recent Enhancements and Use CasesDataWorks Summit
351 views23 slides
Debunking the Myths of HDFS Erasure Coding Performance by
Debunking the Myths of HDFS Erasure Coding Performance Debunking the Myths of HDFS Erasure Coding Performance
Debunking the Myths of HDFS Erasure Coding Performance DataWorks Summit/Hadoop Summit
5.3K views39 slides
HDFS Issues by
HDFS IssuesHDFS Issues
HDFS IssuesSteve Loughran
1.9K views13 slides
Apache ignite as in-memory computing platform by
Apache ignite as in-memory computing platformApache ignite as in-memory computing platform
Apache ignite as in-memory computing platformSurinder Mehra
138 views32 slides

What's hot(20)

Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T... by Spark Summit
Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T...Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T...
Lessons Learned from Dockerizing Spark Workloads: Spark Summit East talk by T...
Spark Summit2K views
Burst Presto & Spark workloads to AWS EMR with no data copies by Alluxio, Inc.
Burst Presto & Spark workloads to AWS EMR with no data copiesBurst Presto & Spark workloads to AWS EMR with no data copies
Burst Presto & Spark workloads to AWS EMR with no data copies
Alluxio, Inc.1.8K views
CaffeOnSpark Update: Recent Enhancements and Use Cases by DataWorks Summit
CaffeOnSpark Update: Recent Enhancements and Use CasesCaffeOnSpark Update: Recent Enhancements and Use Cases
CaffeOnSpark Update: Recent Enhancements and Use Cases
DataWorks Summit351 views
Apache ignite as in-memory computing platform by Surinder Mehra
Apache ignite as in-memory computing platformApache ignite as in-memory computing platform
Apache ignite as in-memory computing platform
Surinder Mehra138 views
HDFS Internals by Apache Apex
HDFS InternalsHDFS Internals
HDFS Internals
Apache Apex2.5K views
Scalable and High available Distributed File System Metadata Service Using gR... by Alluxio, Inc.
Scalable and High available Distributed File System Metadata Service Using gR...Scalable and High available Distributed File System Metadata Service Using gR...
Scalable and High available Distributed File System Metadata Service Using gR...
Alluxio, Inc.2.6K views
Leverage Mesos for running Spark Streaming production jobs by Iulian Dragos a... by Spark Summit
Leverage Mesos for running Spark Streaming production jobs by Iulian Dragos a...Leverage Mesos for running Spark Streaming production jobs by Iulian Dragos a...
Leverage Mesos for running Spark Streaming production jobs by Iulian Dragos a...
Spark Summit2K views
Designing your SaaS Database for Scale with Postgres by Ozgun Erdogan
Designing your SaaS Database for Scale with PostgresDesigning your SaaS Database for Scale with Postgres
Designing your SaaS Database for Scale with Postgres
Ozgun Erdogan539 views
Beyond unit tests: Deployment and testing for Hadoop/Spark workflows by DataWorks Summit
Beyond unit tests: Deployment and testing for Hadoop/Spark workflowsBeyond unit tests: Deployment and testing for Hadoop/Spark workflows
Beyond unit tests: Deployment and testing for Hadoop/Spark workflows
DataWorks Summit781 views
Spark and Object Stores —What You Need to Know: Spark Summit East talk by Ste... by Spark Summit
Spark and Object Stores —What You Need to Know: Spark Summit East talk by Ste...Spark and Object Stores —What You Need to Know: Spark Summit East talk by Ste...
Spark and Object Stores —What You Need to Know: Spark Summit East talk by Ste...
Spark Summit2.9K views
2011 06-30-hadoop-summit v5 by Samuel Rash
2011 06-30-hadoop-summit v52011 06-30-hadoop-summit v5
2011 06-30-hadoop-summit v5
Samuel Rash28K views
HDFS Erasure Code Storage - Same Reliability at Better Storage Efficiency by DataWorks Summit
HDFS Erasure Code Storage - Same Reliability at Better Storage EfficiencyHDFS Erasure Code Storage - Same Reliability at Better Storage Efficiency
HDFS Erasure Code Storage - Same Reliability at Better Storage Efficiency
DataWorks Summit3.3K views
700 Updatable Queries Per Second: Spark as a Real-Time Web Service by Evan Chan
700 Updatable Queries Per Second: Spark as a Real-Time Web Service700 Updatable Queries Per Second: Spark as a Real-Time Web Service
700 Updatable Queries Per Second: Spark as a Real-Time Web Service
Evan Chan993 views
Top 5 mistakes when writing Spark applications by hadooparchbook
Top 5 mistakes when writing Spark applicationsTop 5 mistakes when writing Spark applications
Top 5 mistakes when writing Spark applications
hadooparchbook14.6K views
Hive, Presto, and Spark on TPC-DS benchmark by Dongwon Kim
Hive, Presto, and Spark on TPC-DS benchmarkHive, Presto, and Spark on TPC-DS benchmark
Hive, Presto, and Spark on TPC-DS benchmark
Dongwon Kim9.6K views

Similar to HUG Nov 2010: HDFS Raid - Facebook

Hdfs architecture by
Hdfs architectureHdfs architecture
Hdfs architectureAisha Siddiqa
576 views16 slides
Less is More: 2X Storage Efficiency with HDFS Erasure Coding by
Less is More: 2X Storage Efficiency with HDFS Erasure CodingLess is More: 2X Storage Efficiency with HDFS Erasure Coding
Less is More: 2X Storage Efficiency with HDFS Erasure CodingZhe Zhang
437 views28 slides
Facebook's Approach to Big Data Storage Challenge by
Facebook's Approach to Big Data Storage ChallengeFacebook's Approach to Big Data Storage Challenge
Facebook's Approach to Big Data Storage ChallengeDataWorks Summit
3.6K views27 slides
Hadoop HDFS Concepts by
Hadoop HDFS ConceptsHadoop HDFS Concepts
Hadoop HDFS ConceptsProTechSkills Training
2.2K views21 slides
Container Storage Best Practices in 2017 by
Container Storage Best Practices in 2017Container Storage Best Practices in 2017
Container Storage Best Practices in 2017Keith Resar
3K views40 slides
Raid Levels Technology by
Raid Levels TechnologyRaid Levels Technology
Raid Levels TechnologyIshwor Panta
48 views16 slides

Similar to HUG Nov 2010: HDFS Raid - Facebook(20)

Less is More: 2X Storage Efficiency with HDFS Erasure Coding by Zhe Zhang
Less is More: 2X Storage Efficiency with HDFS Erasure CodingLess is More: 2X Storage Efficiency with HDFS Erasure Coding
Less is More: 2X Storage Efficiency with HDFS Erasure Coding
Zhe Zhang437 views
Facebook's Approach to Big Data Storage Challenge by DataWorks Summit
Facebook's Approach to Big Data Storage ChallengeFacebook's Approach to Big Data Storage Challenge
Facebook's Approach to Big Data Storage Challenge
DataWorks Summit3.6K views
Container Storage Best Practices in 2017 by Keith Resar
Container Storage Best Practices in 2017Container Storage Best Practices in 2017
Container Storage Best Practices in 2017
Keith Resar3K views
Hadoop-professional-software-development-course-in-mumbai by Unmesh Baile
Hadoop-professional-software-development-course-in-mumbaiHadoop-professional-software-development-course-in-mumbai
Hadoop-professional-software-development-course-in-mumbai
Unmesh Baile148 views
Hadoop professional-software-development-course-in-mumbai by Unmesh Baile
Hadoop professional-software-development-course-in-mumbaiHadoop professional-software-development-course-in-mumbai
Hadoop professional-software-development-course-in-mumbai
Unmesh Baile302 views
Hadoop Interview Questions And Answers Part-1 | Big Data Interview Questions ... by Simplilearn
Hadoop Interview Questions And Answers Part-1 | Big Data Interview Questions ...Hadoop Interview Questions And Answers Part-1 | Big Data Interview Questions ...
Hadoop Interview Questions And Answers Part-1 | Big Data Interview Questions ...
Simplilearn734 views
Hadoop training in hyderabad-kellytechnologies by Kelly Technologies
Hadoop training in hyderabad-kellytechnologiesHadoop training in hyderabad-kellytechnologies
Hadoop training in hyderabad-kellytechnologies
Kelly Technologies289 views
Zettabyte File Storage System by Amdocs
Zettabyte File Storage SystemZettabyte File Storage System
Zettabyte File Storage System
Amdocs10.4K views
Zettabyte File Storage System by Amdocs
Zettabyte File Storage SystemZettabyte File Storage System
Zettabyte File Storage System
Amdocs1.4K views
Efficient Shared Data in Perl by Perrin Harkins
Efficient Shared Data in PerlEfficient Shared Data in Perl
Efficient Shared Data in Perl
Perrin Harkins3.7K views

More from Yahoo Developer Network

Developing Mobile Apps for Performance - Swapnil Patel, Verizon Media by
Developing Mobile Apps for Performance - Swapnil Patel, Verizon MediaDeveloping Mobile Apps for Performance - Swapnil Patel, Verizon Media
Developing Mobile Apps for Performance - Swapnil Patel, Verizon MediaYahoo Developer Network
836 views21 slides
Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras... by
Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras...Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras...
Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras...Yahoo Developer Network
561 views20 slides
Athenz & SPIFFE, Tatsuya Yano, Yahoo Japan by
Athenz & SPIFFE, Tatsuya Yano, Yahoo JapanAthenz & SPIFFE, Tatsuya Yano, Yahoo Japan
Athenz & SPIFFE, Tatsuya Yano, Yahoo JapanYahoo Developer Network
614 views29 slides
Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat... by
Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat...Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat...
Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat...Yahoo Developer Network
558 views31 slides
CICD at Oath using Screwdriver by
CICD at Oath using ScrewdriverCICD at Oath using Screwdriver
CICD at Oath using ScrewdriverYahoo Developer Network
884 views8 slides
Big Data Serving with Vespa - Jon Bratseth, Distinguished Architect, Oath by
Big Data Serving with Vespa - Jon Bratseth, Distinguished Architect, OathBig Data Serving with Vespa - Jon Bratseth, Distinguished Architect, Oath
Big Data Serving with Vespa - Jon Bratseth, Distinguished Architect, OathYahoo Developer Network
790 views19 slides

More from Yahoo Developer Network(20)

Developing Mobile Apps for Performance - Swapnil Patel, Verizon Media by Yahoo Developer Network
Developing Mobile Apps for Performance - Swapnil Patel, Verizon MediaDeveloping Mobile Apps for Performance - Swapnil Patel, Verizon Media
Developing Mobile Apps for Performance - Swapnil Patel, Verizon Media
Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras... by Yahoo Developer Network
Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras...Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras...
Athenz - The Open-Source Solution to Provide Access Control in Dynamic Infras...
Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat... by Yahoo Developer Network
Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat...Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat...
Athenz with Istio - Single Access Control Model in Cloud Infrastructures, Tat...
Big Data Serving with Vespa - Jon Bratseth, Distinguished Architect, Oath by Yahoo Developer Network
Big Data Serving with Vespa - Jon Bratseth, Distinguished Architect, OathBig Data Serving with Vespa - Jon Bratseth, Distinguished Architect, Oath
Big Data Serving with Vespa - Jon Bratseth, Distinguished Architect, Oath
How @TwitterHadoop Chose Google Cloud, Joep Rottinghuis, Lohit VijayaRenu by Yahoo Developer Network
How @TwitterHadoop Chose Google Cloud, Joep Rottinghuis, Lohit VijayaRenuHow @TwitterHadoop Chose Google Cloud, Joep Rottinghuis, Lohit VijayaRenu
How @TwitterHadoop Chose Google Cloud, Joep Rottinghuis, Lohit VijayaRenu
The Future of Hadoop in an AI World, Milind Bhandarkar, CEO, Ampool by Yahoo Developer Network
The Future of Hadoop in an AI World, Milind Bhandarkar, CEO, AmpoolThe Future of Hadoop in an AI World, Milind Bhandarkar, CEO, Ampool
The Future of Hadoop in an AI World, Milind Bhandarkar, CEO, Ampool
Apache YARN Federation and Tez at Microsoft, Anupam Upadhyay, Adrian Nicoara,... by Yahoo Developer Network
Apache YARN Federation and Tez at Microsoft, Anupam Upadhyay, Adrian Nicoara,...Apache YARN Federation and Tez at Microsoft, Anupam Upadhyay, Adrian Nicoara,...
Apache YARN Federation and Tez at Microsoft, Anupam Upadhyay, Adrian Nicoara,...
Containerized Services on Apache Hadoop YARN: Past, Present, and Future, Shan... by Yahoo Developer Network
Containerized Services on Apache Hadoop YARN: Past, Present, and Future, Shan...Containerized Services on Apache Hadoop YARN: Past, Present, and Future, Shan...
Containerized Services on Apache Hadoop YARN: Past, Present, and Future, Shan...
HDFS Scalability and Security, Daryn Sharp, Senior Engineer, Oath by Yahoo Developer Network
HDFS Scalability and Security, Daryn Sharp, Senior Engineer, OathHDFS Scalability and Security, Daryn Sharp, Senior Engineer, Oath
HDFS Scalability and Security, Daryn Sharp, Senior Engineer, Oath
Hadoop {Submarine} Project: Running deep learning workloads on YARN, Wangda T... by Yahoo Developer Network
Hadoop {Submarine} Project: Running deep learning workloads on YARN, Wangda T...Hadoop {Submarine} Project: Running deep learning workloads on YARN, Wangda T...
Hadoop {Submarine} Project: Running deep learning workloads on YARN, Wangda T...
Introduction to Vespa – The Open Source Big Data Serving Engine, Jon Bratseth... by Yahoo Developer Network
Introduction to Vespa – The Open Source Big Data Serving Engine, Jon Bratseth...Introduction to Vespa – The Open Source Big Data Serving Engine, Jon Bratseth...
Introduction to Vespa – The Open Source Big Data Serving Engine, Jon Bratseth...
Jun 2017 HUG: Large-Scale Machine Learning: Use Cases and Technologies by Yahoo Developer Network
Jun 2017 HUG: Large-Scale Machine Learning: Use Cases and Technologies Jun 2017 HUG: Large-Scale Machine Learning: Use Cases and Technologies
Jun 2017 HUG: Large-Scale Machine Learning: Use Cases and Technologies
February 2017 HUG: Slow, Stuck, or Runaway Apps? Learn How to Quickly Fix Pro... by Yahoo Developer Network
February 2017 HUG: Slow, Stuck, or Runaway Apps? Learn How to Quickly Fix Pro...February 2017 HUG: Slow, Stuck, or Runaway Apps? Learn How to Quickly Fix Pro...
February 2017 HUG: Slow, Stuck, or Runaway Apps? Learn How to Quickly Fix Pro...
February 2017 HUG: Exactly-once end-to-end processing with Apache Apex by Yahoo Developer Network
February 2017 HUG: Exactly-once end-to-end processing with Apache ApexFebruary 2017 HUG: Exactly-once end-to-end processing with Apache Apex
February 2017 HUG: Exactly-once end-to-end processing with Apache Apex
February 2017 HUG: Data Sketches: A required toolkit for Big Data Analytics by Yahoo Developer Network
February 2017 HUG: Data Sketches: A required toolkit for Big Data AnalyticsFebruary 2017 HUG: Data Sketches: A required toolkit for Big Data Analytics
February 2017 HUG: Data Sketches: A required toolkit for Big Data Analytics

Recently uploaded

SUPPLIER SOURCING.pptx by
SUPPLIER SOURCING.pptxSUPPLIER SOURCING.pptx
SUPPLIER SOURCING.pptxangelicacueva6
15 views1 slide
Kyo - Functional Scala 2023.pdf by
Kyo - Functional Scala 2023.pdfKyo - Functional Scala 2023.pdf
Kyo - Functional Scala 2023.pdfFlavio W. Brasil
368 views92 slides
Zero to Automated in Under a Year by
Zero to Automated in Under a YearZero to Automated in Under a Year
Zero to Automated in Under a YearNetwork Automation Forum
15 views23 slides
Transcript: The Details of Description Techniques tips and tangents on altern... by
Transcript: The Details of Description Techniques tips and tangents on altern...Transcript: The Details of Description Techniques tips and tangents on altern...
Transcript: The Details of Description Techniques tips and tangents on altern...BookNet Canada
136 views15 slides
Five Things You SHOULD Know About Postman by
Five Things You SHOULD Know About PostmanFive Things You SHOULD Know About Postman
Five Things You SHOULD Know About PostmanPostman
33 views43 slides
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N... by
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...James Anderson
85 views32 slides

Recently uploaded(20)

Transcript: The Details of Description Techniques tips and tangents on altern... by BookNet Canada
Transcript: The Details of Description Techniques tips and tangents on altern...Transcript: The Details of Description Techniques tips and tangents on altern...
Transcript: The Details of Description Techniques tips and tangents on altern...
BookNet Canada136 views
Five Things You SHOULD Know About Postman by Postman
Five Things You SHOULD Know About PostmanFive Things You SHOULD Know About Postman
Five Things You SHOULD Know About Postman
Postman33 views
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N... by James Anderson
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
James Anderson85 views
AMAZON PRODUCT RESEARCH.pdf by JerikkLaureta
AMAZON PRODUCT RESEARCH.pdfAMAZON PRODUCT RESEARCH.pdf
AMAZON PRODUCT RESEARCH.pdf
JerikkLaureta26 views
iSAQB Software Architecture Gathering 2023: How Process Orchestration Increas... by Bernd Ruecker
iSAQB Software Architecture Gathering 2023: How Process Orchestration Increas...iSAQB Software Architecture Gathering 2023: How Process Orchestration Increas...
iSAQB Software Architecture Gathering 2023: How Process Orchestration Increas...
Bernd Ruecker37 views
Special_edition_innovator_2023.pdf by WillDavies22
Special_edition_innovator_2023.pdfSpecial_edition_innovator_2023.pdf
Special_edition_innovator_2023.pdf
WillDavies2217 views
TouchLog: Finger Micro Gesture Recognition Using Photo-Reflective Sensors by sugiuralab
TouchLog: Finger Micro Gesture Recognition  Using Photo-Reflective SensorsTouchLog: Finger Micro Gesture Recognition  Using Photo-Reflective Sensors
TouchLog: Finger Micro Gesture Recognition Using Photo-Reflective Sensors
sugiuralab19 views

HUG Nov 2010: HDFS Raid - Facebook

  • 1.
  • 2.
  • 3.
  • 4.
  • 5.
  • 6.
  • 7.
  • 8.
  • 9.
  • 10. Fix missing blocks
  • 11.
  • 12.
  • 13.
  • 14.
  • 15.
  • 16.
  • 17.
  • 18.
  • 19.
  • 20.
  • 21.
  • 22.
  • 23.
  • 24.