Cloud Connect 2012, Big Data @ Netflix

Jerome Boulon
Jerome BoulonCEO/Founder at CaliStream, Data Elite
Big Data @

         Using Big Data to Grow our Business
               & Retain our Customers.

                         Jerome Boulon
           Lead Architect, Hadoop Big Data Infrastructure

                         February 15, 2012
jboulon@netflix.com
Big Data @ Netflix
Offline analysis:
•  Honu: Scalable log analysis system to gain business
   insights:
   –  Errors logs (unstructured logs)
   –  Statistical logs & Performance logs
   –  Etc

Online analysis:
•  Cassandra for all online activities and user facing
   data
   –  A/B testing (test allocation, metadata)
   –  Service level Configuration
   –  etc
                                  2
Overview
                            Data collection pipeline


Applica'on	
                                           Collectors	
  




                 Hive	
                                    M/R	
  




                            Data processing pipeline
                                       3
Honu - Structured Log API
Using	
  Annota+ons	
           Using the Key/Value API
•  Convert Java Class to Hive   •  Produce the same result as
   Table dynamically               Annotation
•  Add/Remove column            •  Avoid unnecessary object
•  Supported java types:           creation
        •  All primitives       •  Fully dynamic
        •  Map                  •  Thread Safe
        •  Object using the
           toString method
Honu, What you get:




log.logEvent(myObject)
                                        Hive table
                         movieId customerId timestamp hostname



      Select customerId, count(1) from MyTable group by customerId;
December 2009
                                                                          Collectors	
  
–    POC for Streaming analysis                Applica'on	
  
–    Single AWS zone
–    1 application
–    60 Millions events/Day
–    50 clients
–    Small Hadoop cluster         Oracle	
  
–    1 Map/Reduce
–    1 Table
                                                                M/R	
  
Feb 2012
                                                40+ Billion events/Day
                                                 8+ tables with 1+TB/Day
                                                100+ smaller tables
                                                Self-serve:
                                                à No DBA
                                                à No Pre-provisioning


                                 	
            	
  
                                                à Fully integrated with Hive
- Multi Regions deployments
- Transparent to our engineers
- Streaming based solution
- Zero configuration
- 7000+ clients
- Built-in:                                           Netflix Hive warehouse
  - Fail-Over
  - Load balancing
                                        	
       à One central Data warehouse
                                                 à Hourly/Daily reports
                                                 à Data retention/expiration
Traceability & Performance
              analysis
•  Track service level call
   –  Instrument low level HTTP client
   –  Calls graph
   –  Request processing vs Perceive latency
   –  Payload marshalling/unmarshalling
      - duration, size, etc
   –  Service Result
      - Status, Error code, Exception, etc
Diagnostic Information
•  Collect latency information for all external
   operations
•  If Latency > threshold log to Honu:
    –  AWS Region & Zone
    –  Instance
    –  Service details
•  Open Jira/Ticket & Attach diagnostic info
Mix Offline and Online Data
Offline data                             Specific conditions
- Fire & forget                          - Online Data availability is not mandatory
- Scale to very large volumes            - If exist, data could be useful online
- Cost effective                         - Only a subset useful Online
                                         - Ready to pay a little bit more




 Special collectors                        Customer support
 - All data goes to Hive                   - Browsing history
 - A subset goes to a real-time system     - Historical & non-critical actions
 - Still cost effective                    Debug
                                           - Push validation
                                           - Root cause analysis
Honu Realtime usages
•  Movie playback experience        •  Customer Support
   –  Video quality                     –  Historical usage
   –  Network issue                     –  Last activity



•  Errors Summary                   •  Launch Reports
    –  Error tracking per service       –  Push validation
    –  Error tracking per device        –  Root cause analysis
Honu Realtime - Architecture
                 Realtime Data collection pipeline


Applica'on	
                                         Collectors	
  




   Real'me	
  
    Access	
  
                             Realtime
                             System                         M/R	
  
A/B Testing
 Test: An experiment where several
 competing behaviors are
 implemented and compared.

 Cell: different experiences within a
 test that are being compared against
 each other.

 Allocation: a customer-specific
 assignment to a cell within a test

Online data:                            Tracking       1 M customers per Test
- Cell Allocation > 1 Billion records   information    8 tracking events per Day
- Test config: 1 entry/test/customer    (example)     ------------------------------------
                                        100 Tests =   800 M events/ Day
                                        3 Months =       72 B events
Movie Presentation A/B Test
A/B Testing - Architecture
         Online Data            Offline Data




- Customer test allocation   - Test tracking
- Metadata about the test    Ex:
Ex:                          - Retention
- Start/End date             - Engagement metrics
- UI directives
- Logging directives
Beacon Server

User behavior
- Client side interactions
- Search/Play/Stop/Pause
                                           Ajax calls
Device monitoring
- Heartbeat
- Status & Key metrics        Beacon	
      Beacon	
     Beacon	
  
BI Integration
Three main technologies

•  Teradata (Data center)
•  Hive (Cloud)
•  Cassandra (Cloud)
Hive ß à BI
–  Dimension tables (daily export from Teradata)
–  Hourly/Daily Hive summary queries
–  Hourly/Daily export from Hive to BI
  •  Queries runs in the cloud
  •  Aggregated result goes back to our BI solution
Hive Reports
Cassandra à BI

•  Use Cassandra backups to run analytics
•  Export SSTable to Hadoop
•  Pig to:
  –  Parse SSTable
  –  Extract/Group required information
•  Load the result back to Teradata
jboulon@gmail.com
www.linkedin.com/in/jboulon
1 of 21

Recommended

Real-time Distributed Stream Processing @ Scale by
Real-time Distributed Stream Processing@ ScaleReal-time Distributed Stream Processing@ Scale
Real-time Distributed Stream Processing @ ScaleJerome Boulon
973 views25 slides
The evolution of the big data platform @ Netflix (OSCON 2015) by
The evolution of the big data platform @ Netflix (OSCON 2015)The evolution of the big data platform @ Netflix (OSCON 2015)
The evolution of the big data platform @ Netflix (OSCON 2015)Eva Tse
51.2K views53 slides
Netflix Big Data Paris 2017 by
Netflix Big Data Paris 2017Netflix Big Data Paris 2017
Netflix Big Data Paris 2017Jason Flittner
494 views25 slides
Data to Drive Decision-Making - CaliStream Meetup by
Data to Drive Decision-Making - CaliStream MeetupData to Drive Decision-Making - CaliStream Meetup
Data to Drive Decision-Making - CaliStream MeetupJerome Boulon
1.4K views36 slides
Lessons Learned - Monitoring the Data Pipeline at Hulu by
Lessons Learned - Monitoring the Data Pipeline at HuluLessons Learned - Monitoring the Data Pipeline at Hulu
Lessons Learned - Monitoring the Data Pipeline at HuluDataWorks Summit
5.2K views42 slides
Big Data Pipeline and Analytics Platform by
Big Data Pipeline and Analytics PlatformBig Data Pipeline and Analytics Platform
Big Data Pipeline and Analytics PlatformSudhir Tonse
1.4K views74 slides

More Related Content

What's hot

ASPgems - kappa architecture by
ASPgems - kappa architectureASPgems - kappa architecture
ASPgems - kappa architectureJuantomás García Molina
2.6K views32 slides
How Disney+ uses fast data ubiquity to improve the customer experience by
 How Disney+ uses fast data ubiquity to improve the customer experience  How Disney+ uses fast data ubiquity to improve the customer experience
How Disney+ uses fast data ubiquity to improve the customer experience Martin Zapletal
207 views34 slides
Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr... by
Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr...Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr...
Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr...Spark Summit
4.1K views24 slides
Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me... by
Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me...Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me...
Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me...HostedbyConfluent
1.6K views21 slides
Headaches and Breakthroughs in Building Continuous Applications by
Headaches and Breakthroughs in Building Continuous ApplicationsHeadaches and Breakthroughs in Building Continuous Applications
Headaches and Breakthroughs in Building Continuous ApplicationsDatabricks
678 views44 slides
Real Time Data Infrastructure team overview by
Real Time Data Infrastructure team overviewReal Time Data Infrastructure team overview
Real Time Data Infrastructure team overviewMonal Daxini
952 views6 slides

What's hot(20)

How Disney+ uses fast data ubiquity to improve the customer experience by Martin Zapletal
 How Disney+ uses fast data ubiquity to improve the customer experience  How Disney+ uses fast data ubiquity to improve the customer experience
How Disney+ uses fast data ubiquity to improve the customer experience
Martin Zapletal207 views
Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr... by Spark Summit
Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr...Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr...
Visualizing AutoTrader Traffic in Near Real-Time with Spark Streaming-(Jon Gr...
Spark Summit4.1K views
Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me... by HostedbyConfluent
Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me...Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me...
Druid + Kafka: transform your data-in-motion to analytics-in-motion | Gian Me...
HostedbyConfluent1.6K views
Headaches and Breakthroughs in Building Continuous Applications by Databricks
Headaches and Breakthroughs in Building Continuous ApplicationsHeadaches and Breakthroughs in Building Continuous Applications
Headaches and Breakthroughs in Building Continuous Applications
Databricks678 views
Real Time Data Infrastructure team overview by Monal Daxini
Real Time Data Infrastructure team overviewReal Time Data Infrastructure team overview
Real Time Data Infrastructure team overview
Monal Daxini952 views
Druid Overview by Rachel Pedreschi by Brian Olsen
Druid Overview by Rachel PedreschiDruid Overview by Rachel Pedreschi
Druid Overview by Rachel Pedreschi
Brian Olsen2.5K views
Apache Spark in Scientific Applciations by Dr. Mirko Kämpf
Apache Spark in Scientific ApplciationsApache Spark in Scientific Applciations
Apache Spark in Scientific Applciations
Dr. Mirko Kämpf380 views
AWS re:Invent 2016| GAM301 | How EA Leveraged Amazon Redshift and AWS Partner... by Amazon Web Services
AWS re:Invent 2016| GAM301 | How EA Leveraged Amazon Redshift and AWS Partner...AWS re:Invent 2016| GAM301 | How EA Leveraged Amazon Redshift and AWS Partner...
AWS re:Invent 2016| GAM301 | How EA Leveraged Amazon Redshift and AWS Partner...
Amazon Web Services1.9K views
The Netflix data platform: Now and in the future by Kurt Brown by Data Con LA
The Netflix data platform: Now and in the future by Kurt BrownThe Netflix data platform: Now and in the future by Kurt Brown
The Netflix data platform: Now and in the future by Kurt Brown
Data Con LA2.8K views
Fast data for fitness 10 nov 2020 by Timothy Spann
Fast data for fitness 10 nov 2020Fast data for fitness 10 nov 2020
Fast data for fitness 10 nov 2020
Timothy Spann395 views
Netflix incloudsmarch8 2011forwiki by Kevin McEntee
Netflix incloudsmarch8 2011forwikiNetflix incloudsmarch8 2011forwiki
Netflix incloudsmarch8 2011forwiki
Kevin McEntee186.2K views
Getting It Right Exactly Once: Principles for Streaming Architectures by SingleStore
Getting It Right Exactly Once: Principles for Streaming ArchitecturesGetting It Right Exactly Once: Principles for Streaming Architectures
Getting It Right Exactly Once: Principles for Streaming Architectures
SingleStore2.4K views
Processing Real-Time Data at Scale: A streaming platform as a central nervous... by confluent
Processing Real-Time Data at Scale: A streaming platform as a central nervous...Processing Real-Time Data at Scale: A streaming platform as a central nervous...
Processing Real-Time Data at Scale: A streaming platform as a central nervous...
confluent394 views
Spark at Airbnb by Hao Wang
Spark at AirbnbSpark at Airbnb
Spark at Airbnb
Hao Wang545 views
A unified analytics platform with Kafka and Flink | Stephan Ewen, Ververica by HostedbyConfluent
A unified analytics platform with Kafka and Flink | Stephan Ewen, VervericaA unified analytics platform with Kafka and Flink | Stephan Ewen, Ververica
A unified analytics platform with Kafka and Flink | Stephan Ewen, Ververica
Cloud-Based Event Stream Processing Architectures and Patterns with Apache Ka... by HostedbyConfluent
Cloud-Based Event Stream Processing Architectures and Patterns with Apache Ka...Cloud-Based Event Stream Processing Architectures and Patterns with Apache Ka...
Cloud-Based Event Stream Processing Architectures and Patterns with Apache Ka...
HostedbyConfluent740 views
Data in Motion: Building Stream-Based Architectures with Qlik Replicate & Kaf... by HostedbyConfluent
Data in Motion: Building Stream-Based Architectures with Qlik Replicate & Kaf...Data in Motion: Building Stream-Based Architectures with Qlik Replicate & Kaf...
Data in Motion: Building Stream-Based Architectures with Qlik Replicate & Kaf...
HostedbyConfluent1.4K views
Big Data Kappa | Mark Senerth, The Walt Disney Company - DMED, Data Tech by HostedbyConfluent
Big Data Kappa | Mark Senerth, The Walt Disney Company - DMED, Data TechBig Data Kappa | Mark Senerth, The Walt Disney Company - DMED, Data Tech
Big Data Kappa | Mark Senerth, The Walt Disney Company - DMED, Data Tech
HostedbyConfluent2.5K views

Viewers also liked

The Netflix Way to deal with Big Data Problems by
The Netflix Way to deal with Big Data ProblemsThe Netflix Way to deal with Big Data Problems
The Netflix Way to deal with Big Data ProblemsMonal Daxini
5K views82 slides
The Big Data TV: Data Analytics, Algorithm, and Netflix’s Original Programming by
The Big Data TV: Data Analytics, Algorithm, and Netflix’s Original ProgrammingThe Big Data TV: Data Analytics, Algorithm, and Netflix’s Original Programming
The Big Data TV: Data Analytics, Algorithm, and Netflix’s Original Programminghye-jin-lee
3.6K views6 slides
ACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILES by
ACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILESACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILES
ACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILESAdrija Chowdhury
16.5K views25 slides
Presto @ Netflix: Interactive Queries at Petabyte Scale by
Presto @ Netflix: Interactive Queries at Petabyte ScalePresto @ Netflix: Interactive Queries at Petabyte Scale
Presto @ Netflix: Interactive Queries at Petabyte ScaleDataWorks Summit
1.6K views31 slides
Mobile Phone Based Drunk driving detection by
Mobile Phone Based Drunk driving detectionMobile Phone Based Drunk driving detection
Mobile Phone Based Drunk driving detectionnagarajc007
3.8K views17 slides
Netflix - Enabling a Culture of Analytics by
Netflix - Enabling a Culture of AnalyticsNetflix - Enabling a Culture of Analytics
Netflix - Enabling a Culture of AnalyticsBlake Irvine
22.4K views19 slides

Viewers also liked(6)

The Netflix Way to deal with Big Data Problems by Monal Daxini
The Netflix Way to deal with Big Data ProblemsThe Netflix Way to deal with Big Data Problems
The Netflix Way to deal with Big Data Problems
Monal Daxini5K views
The Big Data TV: Data Analytics, Algorithm, and Netflix’s Original Programming by hye-jin-lee
The Big Data TV: Data Analytics, Algorithm, and Netflix’s Original ProgrammingThe Big Data TV: Data Analytics, Algorithm, and Netflix’s Original Programming
The Big Data TV: Data Analytics, Algorithm, and Netflix’s Original Programming
hye-jin-lee3.6K views
ACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILES by Adrija Chowdhury
ACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILESACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILES
ACCIDENT PREVENTION AND SECURITY SYSTEM FOR AUTOMOBILES
Adrija Chowdhury16.5K views
Presto @ Netflix: Interactive Queries at Petabyte Scale by DataWorks Summit
Presto @ Netflix: Interactive Queries at Petabyte ScalePresto @ Netflix: Interactive Queries at Petabyte Scale
Presto @ Netflix: Interactive Queries at Petabyte Scale
DataWorks Summit1.6K views
Mobile Phone Based Drunk driving detection by nagarajc007
Mobile Phone Based Drunk driving detectionMobile Phone Based Drunk driving detection
Mobile Phone Based Drunk driving detection
nagarajc0073.8K views
Netflix - Enabling a Culture of Analytics by Blake Irvine
Netflix - Enabling a Culture of AnalyticsNetflix - Enabling a Culture of Analytics
Netflix - Enabling a Culture of Analytics
Blake Irvine22.4K views

Similar to Cloud Connect 2012, Big Data @ Netflix

Building Scalable Big Data Infrastructure Using Open Source Software Presenta... by
Building Scalable Big Data Infrastructure Using Open Source Software Presenta...Building Scalable Big Data Infrastructure Using Open Source Software Presenta...
Building Scalable Big Data Infrastructure Using Open Source Software Presenta...ssuserd3a367
11 views37 slides
Unified Framework for Real Time, Near Real Time and Offline Analysis of Video... by
Unified Framework for Real Time, Near Real Time and Offline Analysis of Video...Unified Framework for Real Time, Near Real Time and Offline Analysis of Video...
Unified Framework for Real Time, Near Real Time and Offline Analysis of Video...Spark Summit
1.6K views35 slides
Pulsar: Real-time Analytics at Scale with Kafka, Kylin and Druid by
Pulsar: Real-time Analytics at Scale with Kafka, Kylin and DruidPulsar: Real-time Analytics at Scale with Kafka, Kylin and Druid
Pulsar: Real-time Analytics at Scale with Kafka, Kylin and DruidTony Ng
6.3K views34 slides
Using real time big data analytics for competitive advantage by
 Using real time big data analytics for competitive advantage Using real time big data analytics for competitive advantage
Using real time big data analytics for competitive advantageAmazon Web Services
2.3K views33 slides
Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D... by
Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D...Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D...
Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D...SL Corporation
444 views41 slides
Machine Learning for Smarter Apps - Jacksonville Meetup by
Machine Learning for Smarter Apps - Jacksonville MeetupMachine Learning for Smarter Apps - Jacksonville Meetup
Machine Learning for Smarter Apps - Jacksonville MeetupSri Ambati
1.9K views49 slides

Similar to Cloud Connect 2012, Big Data @ Netflix(20)

Building Scalable Big Data Infrastructure Using Open Source Software Presenta... by ssuserd3a367
Building Scalable Big Data Infrastructure Using Open Source Software Presenta...Building Scalable Big Data Infrastructure Using Open Source Software Presenta...
Building Scalable Big Data Infrastructure Using Open Source Software Presenta...
ssuserd3a36711 views
Unified Framework for Real Time, Near Real Time and Offline Analysis of Video... by Spark Summit
Unified Framework for Real Time, Near Real Time and Offline Analysis of Video...Unified Framework for Real Time, Near Real Time and Offline Analysis of Video...
Unified Framework for Real Time, Near Real Time and Offline Analysis of Video...
Spark Summit1.6K views
Pulsar: Real-time Analytics at Scale with Kafka, Kylin and Druid by Tony Ng
Pulsar: Real-time Analytics at Scale with Kafka, Kylin and DruidPulsar: Real-time Analytics at Scale with Kafka, Kylin and Druid
Pulsar: Real-time Analytics at Scale with Kafka, Kylin and Druid
Tony Ng6.3K views
Using real time big data analytics for competitive advantage by Amazon Web Services
 Using real time big data analytics for competitive advantage Using real time big data analytics for competitive advantage
Using real time big data analytics for competitive advantage
Amazon Web Services2.3K views
Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D... by SL Corporation
Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D...Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D...
Overcoming the Top Four Challenges to Real‐Time Performance in Large‐Scale, D...
SL Corporation444 views
Machine Learning for Smarter Apps - Jacksonville Meetup by Sri Ambati
Machine Learning for Smarter Apps - Jacksonville MeetupMachine Learning for Smarter Apps - Jacksonville Meetup
Machine Learning for Smarter Apps - Jacksonville Meetup
Sri Ambati1.9K views
Processing Big Data by cwensel
Processing Big DataProcessing Big Data
Processing Big Data
cwensel817 views
The Heart of the Data Mesh Beats in Real-Time with Apache Kafka by Kai Wähner
The Heart of the Data Mesh Beats in Real-Time with Apache KafkaThe Heart of the Data Mesh Beats in Real-Time with Apache Kafka
The Heart of the Data Mesh Beats in Real-Time with Apache Kafka
Kai Wähner2K views
Resistance is futile, resilience is crucial by Hristo Iliev
Resistance is futile, resilience is crucialResistance is futile, resilience is crucial
Resistance is futile, resilience is crucial
Hristo Iliev133 views
Introduction to Stream Processing by Guido Schmutz
Introduction to Stream ProcessingIntroduction to Stream Processing
Introduction to Stream Processing
Guido Schmutz1.3K views
How Netflix Monitors Applications in Near Real-time w Amazon Kinesis - ABD401... by Amazon Web Services
How Netflix Monitors Applications in Near Real-time w Amazon Kinesis - ABD401...How Netflix Monitors Applications in Near Real-time w Amazon Kinesis - ABD401...
How Netflix Monitors Applications in Near Real-time w Amazon Kinesis - ABD401...
Amazon Web Services7.8K views
Overcoming the Top Four Challenges to Real-Time Performance in Large-Scale, D... by SL Corporation
Overcoming the Top Four Challenges to Real-Time Performance in Large-Scale, D...Overcoming the Top Four Challenges to Real-Time Performance in Large-Scale, D...
Overcoming the Top Four Challenges to Real-Time Performance in Large-Scale, D...
SL Corporation693 views
Stream processing on mobile networks by pbelko82
Stream processing on mobile networksStream processing on mobile networks
Stream processing on mobile networks
pbelko82311 views
DevOps in the Amazon Cloud – Learn from the pioneersNetflix suro by Gaurav "GP" Pal
DevOps in the Amazon Cloud – Learn from the pioneersNetflix suroDevOps in the Amazon Cloud – Learn from the pioneersNetflix suro
DevOps in the Amazon Cloud – Learn from the pioneersNetflix suro
Gaurav "GP" Pal1.1K views
Combining Hadoop RDBMS for Large-Scale Big Data Analytics by DataWorks Summit
Combining Hadoop RDBMS for Large-Scale Big Data AnalyticsCombining Hadoop RDBMS for Large-Scale Big Data Analytics
Combining Hadoop RDBMS for Large-Scale Big Data Analytics
DataWorks Summit12.1K views
Building real time data-driven products by Lars Albertsson
Building real time data-driven productsBuilding real time data-driven products
Building real time data-driven products
Lars Albertsson2.8K views
Kognitio overview jan 2013 by Kognitio
Kognitio overview jan 2013Kognitio overview jan 2013
Kognitio overview jan 2013
Kognitio124 views

Recently uploaded

CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue by
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlueCloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlueShapeBlue
135 views13 slides
Business Analyst Series 2023 - Week 4 Session 7 by
Business Analyst Series 2023 -  Week 4 Session 7Business Analyst Series 2023 -  Week 4 Session 7
Business Analyst Series 2023 - Week 4 Session 7DianaGray10
139 views31 slides
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And... by
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...ShapeBlue
106 views12 slides
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue by
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlueElevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlueShapeBlue
222 views7 slides
Qualifying SaaS, IaaS.pptx by
Qualifying SaaS, IaaS.pptxQualifying SaaS, IaaS.pptx
Qualifying SaaS, IaaS.pptxSachin Bhandari
1K views8 slides
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit... by
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...ShapeBlue
159 views25 slides

Recently uploaded(20)

CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue by ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlueCloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
ShapeBlue135 views
Business Analyst Series 2023 - Week 4 Session 7 by DianaGray10
Business Analyst Series 2023 -  Week 4 Session 7Business Analyst Series 2023 -  Week 4 Session 7
Business Analyst Series 2023 - Week 4 Session 7
DianaGray10139 views
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And... by ShapeBlue
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...
ShapeBlue106 views
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue by ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlueElevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
ShapeBlue222 views
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit... by ShapeBlue
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...
ShapeBlue159 views
Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ... by ShapeBlue
Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ...Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ...
Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ...
ShapeBlue184 views
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T by ShapeBlue
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&TCloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T
ShapeBlue152 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 Ruecker54 views
2FA and OAuth2 in CloudStack - Andrija Panić - ShapeBlue by ShapeBlue
2FA and OAuth2 in CloudStack - Andrija Panić - ShapeBlue2FA and OAuth2 in CloudStack - Andrija Panić - ShapeBlue
2FA and OAuth2 in CloudStack - Andrija Panić - ShapeBlue
ShapeBlue147 views
Backroll, News and Demo - Pierre Charton, Matthias Dhellin, Ousmane Diarra - ... by ShapeBlue
Backroll, News and Demo - Pierre Charton, Matthias Dhellin, Ousmane Diarra - ...Backroll, News and Demo - Pierre Charton, Matthias Dhellin, Ousmane Diarra - ...
Backroll, News and Demo - Pierre Charton, Matthias Dhellin, Ousmane Diarra - ...
ShapeBlue186 views
Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ... by ShapeBlue
Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ...Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ...
Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ...
ShapeBlue119 views
Why and How CloudStack at weSystems - Stephan Bienek - weSystems by ShapeBlue
Why and How CloudStack at weSystems - Stephan Bienek - weSystemsWhy and How CloudStack at weSystems - Stephan Bienek - weSystems
Why and How CloudStack at weSystems - Stephan Bienek - weSystems
ShapeBlue238 views
Hypervisor Agnostic DRS in CloudStack - Brief overview & demo - Vishesh Jinda... by ShapeBlue
Hypervisor Agnostic DRS in CloudStack - Brief overview & demo - Vishesh Jinda...Hypervisor Agnostic DRS in CloudStack - Brief overview & demo - Vishesh Jinda...
Hypervisor Agnostic DRS in CloudStack - Brief overview & demo - Vishesh Jinda...
ShapeBlue161 views
NTGapps NTG LowCode Platform by Mustafa Kuğu
NTGapps NTG LowCode Platform NTGapps NTG LowCode Platform
NTGapps NTG LowCode Platform
Mustafa Kuğu423 views
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ... by ShapeBlue
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...
ShapeBlue166 views
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ... by ShapeBlue
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...
ShapeBlue126 views
Setting Up Your First CloudStack Environment with Beginners Challenges - MD R... by ShapeBlue
Setting Up Your First CloudStack Environment with Beginners Challenges - MD R...Setting Up Your First CloudStack Environment with Beginners Challenges - MD R...
Setting Up Your First CloudStack Environment with Beginners Challenges - MD R...
ShapeBlue173 views
Keynote Talk: Open Source is Not Dead - Charles Schulz - Vates by ShapeBlue
Keynote Talk: Open Source is Not Dead - Charles Schulz - VatesKeynote Talk: Open Source is Not Dead - Charles Schulz - Vates
Keynote Talk: Open Source is Not Dead - Charles Schulz - Vates
ShapeBlue252 views
DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti... by ShapeBlue
DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti...DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti...
DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti...
ShapeBlue139 views

Cloud Connect 2012, Big Data @ Netflix

  • 1. Big Data @ Using Big Data to Grow our Business & Retain our Customers. Jerome Boulon Lead Architect, Hadoop Big Data Infrastructure February 15, 2012 jboulon@netflix.com
  • 2. Big Data @ Netflix Offline analysis: •  Honu: Scalable log analysis system to gain business insights: –  Errors logs (unstructured logs) –  Statistical logs & Performance logs –  Etc Online analysis: •  Cassandra for all online activities and user facing data –  A/B testing (test allocation, metadata) –  Service level Configuration –  etc 2
  • 3. Overview Data collection pipeline Applica'on   Collectors   Hive   M/R   Data processing pipeline 3
  • 4. Honu - Structured Log API Using  Annota+ons   Using the Key/Value API •  Convert Java Class to Hive •  Produce the same result as Table dynamically Annotation •  Add/Remove column •  Avoid unnecessary object •  Supported java types: creation •  All primitives •  Fully dynamic •  Map •  Thread Safe •  Object using the toString method
  • 5. Honu, What you get: log.logEvent(myObject) Hive table movieId customerId timestamp hostname Select customerId, count(1) from MyTable group by customerId;
  • 6. December 2009 Collectors   –  POC for Streaming analysis Applica'on   –  Single AWS zone –  1 application –  60 Millions events/Day –  50 clients –  Small Hadoop cluster Oracle   –  1 Map/Reduce –  1 Table M/R  
  • 7. Feb 2012 40+ Billion events/Day 8+ tables with 1+TB/Day 100+ smaller tables Self-serve: à No DBA à No Pre-provisioning     à Fully integrated with Hive - Multi Regions deployments - Transparent to our engineers - Streaming based solution - Zero configuration - 7000+ clients - Built-in: Netflix Hive warehouse - Fail-Over - Load balancing   à One central Data warehouse à Hourly/Daily reports à Data retention/expiration
  • 8. Traceability & Performance analysis •  Track service level call –  Instrument low level HTTP client –  Calls graph –  Request processing vs Perceive latency –  Payload marshalling/unmarshalling - duration, size, etc –  Service Result - Status, Error code, Exception, etc
  • 9. Diagnostic Information •  Collect latency information for all external operations •  If Latency > threshold log to Honu: –  AWS Region & Zone –  Instance –  Service details •  Open Jira/Ticket & Attach diagnostic info
  • 10. Mix Offline and Online Data Offline data Specific conditions - Fire & forget - Online Data availability is not mandatory - Scale to very large volumes - If exist, data could be useful online - Cost effective - Only a subset useful Online - Ready to pay a little bit more Special collectors Customer support - All data goes to Hive - Browsing history - A subset goes to a real-time system - Historical & non-critical actions - Still cost effective Debug - Push validation - Root cause analysis
  • 11. Honu Realtime usages •  Movie playback experience •  Customer Support –  Video quality –  Historical usage –  Network issue –  Last activity •  Errors Summary •  Launch Reports –  Error tracking per service –  Push validation –  Error tracking per device –  Root cause analysis
  • 12. Honu Realtime - Architecture Realtime Data collection pipeline Applica'on   Collectors   Real'me   Access   Realtime System M/R  
  • 13. A/B Testing Test: An experiment where several competing behaviors are implemented and compared. Cell: different experiences within a test that are being compared against each other. Allocation: a customer-specific assignment to a cell within a test Online data: Tracking 1 M customers per Test - Cell Allocation > 1 Billion records information 8 tracking events per Day - Test config: 1 entry/test/customer (example) ------------------------------------ 100 Tests = 800 M events/ Day 3 Months = 72 B events
  • 15. A/B Testing - Architecture Online Data Offline Data - Customer test allocation - Test tracking - Metadata about the test Ex: Ex: - Retention - Start/End date - Engagement metrics - UI directives - Logging directives
  • 16. Beacon Server User behavior - Client side interactions - Search/Play/Stop/Pause Ajax calls Device monitoring - Heartbeat - Status & Key metrics Beacon   Beacon   Beacon  
  • 17. BI Integration Three main technologies •  Teradata (Data center) •  Hive (Cloud) •  Cassandra (Cloud)
  • 18. Hive ß à BI –  Dimension tables (daily export from Teradata) –  Hourly/Daily Hive summary queries –  Hourly/Daily export from Hive to BI •  Queries runs in the cloud •  Aggregated result goes back to our BI solution
  • 20. Cassandra à BI •  Use Cassandra backups to run analytics •  Export SSTable to Hadoop •  Pig to: –  Parse SSTable –  Extract/Group required information •  Load the result back to Teradata