SlideShare a Scribd company logo
1 of 19
STREAMING ARCHITECTURE
Xun Liu 10-27-2018
AGENDA
1/6/2021 © 2012 Telenav, Proprietary and Confidential 2
• Story: Build Netflix
• Streaming tool: Kafka
• What we could do?
BUILD NETFLIX
1/6/2021 © 2012 Telenav, Proprietary and Confidential 3
Problems of grey boxes
1. Multiple techniques
for arrow between
different components
2. Dependency between
two components
MESSAGE TECHNOLOGY
1/6/2021 © 2012 Telenav, Proprietary and Confidential 4
Remove unnecessary
dependency and
complexity
MORE REQUIREMENTS FOR THUMBNAIL
1/6/2021 © 2012 Telenav, Proprietary and Confidential 5
MORE ADJUSTMENT FOR THUMBNAIL
1/6/2021 © 2012 Telenav, Proprietary and Confidential 6
FINAL
1/6/2021 © 2012 Telenav, Proprietary and Confidential 7
FINAL
1/6/2021 © 2012 Telenav, Proprietary and Confidential 8
• Universal microservices architecture
• Maintaining independence of microservices by using
lightweight communications between services via a
uniform message-passing technology (such as Apache
Kafka or MapR Streams) that is durable and high
performance
• Use of distributed files, NoSQL databases, and snapshot
WHAT IS KAFKA
1/6/2021 © 2012 Telenav, Proprietary and Confidential 9
WHY KAFKA
• From Jay Kreps
#1. Transporting data between system
#2. Core of Analytics system
• Original problem raised by Jay Kreps: When you use
database to analysis data, during analysis time new
data come in, how to reflect changes? Tails+Tails +
Tails? MessageMQ?
1/6/2021 © 2012 Telenav, Proprietary and Confidential 10
WHY KAFKA
• Message durability(commit log, append only behavior)
• Keep message sequence
• Designed for large volume of data
• One message be consumed by multiple consumer for
different purpose
• Message be consumed Just once for each topic
1/6/2021 © 2012 Telenav, Proprietary and Confidential 11
KAFKA TERMINOLOGY
• Message Broker: serve to persist data and handle
request from both producer and consumer
• Producer client: produce publish
• Consumer client: consume subscribe
• Topic: partition at message broker layer
when have large and large volume of data could
scale up, partition topic
partition: concept of topic, parameters
user specific or generated based on
certain pattern
1/6/2021 © 2012 Telenav, Proprietary and Confidential 12
KAFKA SCENARIOS
• #1. online data to offline data
Online data like page view/impression, -> offline
hadoop system, kafka be consumed by ETL
• #2. online data to online pipeline
Emails/news feed
• #3. Offline data to online pipeline
Data be generated in offline data wareehouse -> online
machine learning
1/6/2021 © 2012 Telenav, Proprietary and Confidential 13
CORE OF KAFKA – REPLICATED DISTRIBUTE LOG
Log is an append-only, totally-ordered sequence of
records ordered by time
1/6/2021 © 2012 Telenav, Proprietary and Confidential 14
WHY LOG IS IMPORTANT
• The answer is that logs have a specific purpose: they
record what happened and when
• The two problems a log solves—ordering changes and
distributing data—are even more important in
distributed data systems.
• One of the beautiful things about this approach is that
the time stamps that index the log now act as the clock
for the state of the replicas—you can describe each
replica by a single number, the timestamp for the
maximum log entry it has processed. This timestamp
combined with the log uniquely captures the entire
state of the replica.
1/6/2021 © 2012 Telenav, Proprietary and Confidential 15
STATE MACHINE REPLICATION
• State-Machine-> active active -> The active-active approach
might log out the transformations to apply, say "+1", "*2", etc.
Each replica would apply these transformations and hence go
through the same set of values.
• Primary-Backup -> Active-Passive -> a single master execute the
transformations and log out the result, say "1", "3", "6", etc.
1/6/2021 © 2012 Telenav, Proprietary and Confidential 16
1/6/2021 © 2012 Telenav, Proprietary and Confidential 17
RELATION WITH TELENAV
• Decouples Data producer with consumer
• Log Replay(A/B test, parallel test)
1/6/2021 © 2012 Telenav, Proprietary and Confidential 18
User Prob Data
User Service
Requset
Data update
Message Bus(Kafka)
Live
data(weather,
traffic)
Data analytics
A/B test
Live traffic
generation
Data
preprocessing
Kafka
Logs
data
data
data
data
REFERENCE
• Kafka Stream Examples
https://github.com/CodeBear801/kafka-streams-
examples
• Jay Kreps https://engineering.linkedin.com/distributed-
systems/log-what-every-software-engineer-should-
know-about-real-time-datas-unifying
https://www.confluent.io/blog/stream-data-platform-1
1/6/2021 © 2012 Telenav, Proprietary and Confidential 19

More Related Content

Similar to Build a Streaming Architecture with Kafka

What’s New in Documentum 7.3
What’s New in Documentum 7.3What’s New in Documentum 7.3
What’s New in Documentum 7.3Michael Mohen
 
A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...
A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...
A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...DaoliCloud Ltd
 
Evolving from Messaging to Event Streaming
Evolving from Messaging to Event StreamingEvolving from Messaging to Event Streaming
Evolving from Messaging to Event Streamingconfluent
 
How eStruxture Data Centers is Using ECE to Rapidly Scale Their Business
How eStruxture Data Centers is Using ECE to Rapidly Scale Their BusinessHow eStruxture Data Centers is Using ECE to Rapidly Scale Their Business
How eStruxture Data Centers is Using ECE to Rapidly Scale Their BusinessElasticsearch
 
Continus sql with sql stream builder
Continus sql with sql stream builderContinus sql with sql stream builder
Continus sql with sql stream builderTimothy Spann
 
Hlb private cloud rules of engagement idc
Hlb private cloud rules of engagement   idcHlb private cloud rules of engagement   idc
Hlb private cloud rules of engagement idcYew Jin Kang
 
Confluent Partner Tech Talk with QLIK
Confluent Partner Tech Talk with QLIKConfluent Partner Tech Talk with QLIK
Confluent Partner Tech Talk with QLIKconfluent
 
SharePoint Best Practices Conference 2013
SharePoint Best Practices Conference 2013SharePoint Best Practices Conference 2013
SharePoint Best Practices Conference 2013Mike Brannon
 
Scaling mature systems
Scaling mature systemsScaling mature systems
Scaling mature systemsHanMorten
 
Large customers want postgresql too !!
Large customers want postgresql too !!Large customers want postgresql too !!
Large customers want postgresql too !!rosensteel
 
InfluxDB Live Product Training
InfluxDB Live Product TrainingInfluxDB Live Product Training
InfluxDB Live Product TrainingInfluxData
 
Introducing Events and Stream Processing into Nationwide Building Society
Introducing Events and Stream Processing into Nationwide Building SocietyIntroducing Events and Stream Processing into Nationwide Building Society
Introducing Events and Stream Processing into Nationwide Building Societyconfluent
 
Bringing Real-Time to the Enterprise with Hortonworks DataFlow
Bringing Real-Time to the Enterprise with Hortonworks DataFlowBringing Real-Time to the Enterprise with Hortonworks DataFlow
Bringing Real-Time to the Enterprise with Hortonworks DataFlowDataWorks Summit
 
Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...
Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...
Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...Phil Wilkins
 
Lunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps Approach
Lunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps ApproachLunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps Approach
Lunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps ApproachDenodo
 
Mastering Cloud Data Cost Control: A FinOps Approach
Mastering Cloud Data Cost Control: A FinOps ApproachMastering Cloud Data Cost Control: A FinOps Approach
Mastering Cloud Data Cost Control: A FinOps ApproachDenodo
 
MuleSoft Online Meetup - MuleSoft integration with snowflake and kafka
MuleSoft Online Meetup - MuleSoft integration with snowflake and kafkaMuleSoft Online Meetup - MuleSoft integration with snowflake and kafka
MuleSoft Online Meetup - MuleSoft integration with snowflake and kafkaRoyston Lobo
 
eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...
eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...
eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...eFolder
 
Modernize Your Banking Platform with Temenos and NuoDB
Modernize Your Banking Platform with Temenos and NuoDBModernize Your Banking Platform with Temenos and NuoDB
Modernize Your Banking Platform with Temenos and NuoDBNuoDB
 
OpenFlow in Enterprise Data Centers - Products, Lessons and Requirements
OpenFlow in Enterprise Data Centers - Products, Lessons and RequirementsOpenFlow in Enterprise Data Centers - Products, Lessons and Requirements
OpenFlow in Enterprise Data Centers - Products, Lessons and RequirementsOpen Networking Summits
 

Similar to Build a Streaming Architecture with Kafka (20)

What’s New in Documentum 7.3
What’s New in Documentum 7.3What’s New in Documentum 7.3
What’s New in Documentum 7.3
 
A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...
A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...
A Novel Use of Openflow and Its Applications in Connecting Docker and Dummify...
 
Evolving from Messaging to Event Streaming
Evolving from Messaging to Event StreamingEvolving from Messaging to Event Streaming
Evolving from Messaging to Event Streaming
 
How eStruxture Data Centers is Using ECE to Rapidly Scale Their Business
How eStruxture Data Centers is Using ECE to Rapidly Scale Their BusinessHow eStruxture Data Centers is Using ECE to Rapidly Scale Their Business
How eStruxture Data Centers is Using ECE to Rapidly Scale Their Business
 
Continus sql with sql stream builder
Continus sql with sql stream builderContinus sql with sql stream builder
Continus sql with sql stream builder
 
Hlb private cloud rules of engagement idc
Hlb private cloud rules of engagement   idcHlb private cloud rules of engagement   idc
Hlb private cloud rules of engagement idc
 
Confluent Partner Tech Talk with QLIK
Confluent Partner Tech Talk with QLIKConfluent Partner Tech Talk with QLIK
Confluent Partner Tech Talk with QLIK
 
SharePoint Best Practices Conference 2013
SharePoint Best Practices Conference 2013SharePoint Best Practices Conference 2013
SharePoint Best Practices Conference 2013
 
Scaling mature systems
Scaling mature systemsScaling mature systems
Scaling mature systems
 
Large customers want postgresql too !!
Large customers want postgresql too !!Large customers want postgresql too !!
Large customers want postgresql too !!
 
InfluxDB Live Product Training
InfluxDB Live Product TrainingInfluxDB Live Product Training
InfluxDB Live Product Training
 
Introducing Events and Stream Processing into Nationwide Building Society
Introducing Events and Stream Processing into Nationwide Building SocietyIntroducing Events and Stream Processing into Nationwide Building Society
Introducing Events and Stream Processing into Nationwide Building Society
 
Bringing Real-Time to the Enterprise with Hortonworks DataFlow
Bringing Real-Time to the Enterprise with Hortonworks DataFlowBringing Real-Time to the Enterprise with Hortonworks DataFlow
Bringing Real-Time to the Enterprise with Hortonworks DataFlow
 
Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...
Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...
Fluentd – Making Logging Easy & Effective in a Multi-cloud & Hybrid Environme...
 
Lunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps Approach
Lunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps ApproachLunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps Approach
Lunch and Learn ANZ: Mastering Cloud Data Cost Control: A FinOps Approach
 
Mastering Cloud Data Cost Control: A FinOps Approach
Mastering Cloud Data Cost Control: A FinOps ApproachMastering Cloud Data Cost Control: A FinOps Approach
Mastering Cloud Data Cost Control: A FinOps Approach
 
MuleSoft Online Meetup - MuleSoft integration with snowflake and kafka
MuleSoft Online Meetup - MuleSoft integration with snowflake and kafkaMuleSoft Online Meetup - MuleSoft integration with snowflake and kafka
MuleSoft Online Meetup - MuleSoft integration with snowflake and kafka
 
eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...
eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...
eFolder Webinar — Special eFolder Announcement: StorageCraft Agreement and CE...
 
Modernize Your Banking Platform with Temenos and NuoDB
Modernize Your Banking Platform with Temenos and NuoDBModernize Your Banking Platform with Temenos and NuoDB
Modernize Your Banking Platform with Temenos and NuoDB
 
OpenFlow in Enterprise Data Centers - Products, Lessons and Requirements
OpenFlow in Enterprise Data Centers - Products, Lessons and RequirementsOpenFlow in Enterprise Data Centers - Products, Lessons and Requirements
OpenFlow in Enterprise Data Centers - Products, Lessons and Requirements
 

Recently uploaded

定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一
定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一
定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一ra6e69ou
 
c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...
c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...
c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...gurkirankumar98700
 
Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...
Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...
Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...baharayali
 
CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service 👖
CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service  👖CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service  👖
CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service 👖anilsa9823
 
O9654467111 Call Girls In Dwarka Women Seeking Men
O9654467111 Call Girls In Dwarka Women Seeking MenO9654467111 Call Girls In Dwarka Women Seeking Men
O9654467111 Call Girls In Dwarka Women Seeking MenSapana Sha
 
Learn About the Rise of Instagram Pro in 2024
Learn About the Rise of Instagram Pro in 2024Learn About the Rise of Instagram Pro in 2024
Learn About the Rise of Instagram Pro in 2024Islam Fit
 
Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779
Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779
Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779Delhi Call girls
 
Spotify AI DJ Deck - The Agency at University of Florida
Spotify AI DJ Deck - The Agency at University of FloridaSpotify AI DJ Deck - The Agency at University of Florida
Spotify AI DJ Deck - The Agency at University of Floridajorirz24
 
Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779
Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779
Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779Delhi Call girls
 
Call Girls In Patel Nagar Delhi 9654467111 Escorts Service
Call Girls In Patel Nagar Delhi 9654467111 Escorts ServiceCall Girls In Patel Nagar Delhi 9654467111 Escorts Service
Call Girls In Patel Nagar Delhi 9654467111 Escorts ServiceSapana Sha
 
Night 7k Call Girls Noida Sector 120 Call Me: 8448380779
Night 7k Call Girls Noida Sector 120 Call Me: 8448380779Night 7k Call Girls Noida Sector 120 Call Me: 8448380779
Night 7k Call Girls Noida Sector 120 Call Me: 8448380779Delhi Call girls
 
Top Call Girls In Charbagh ( Lucknow ) 🔝 8923113531 🔝 Cash Payment
Top Call Girls In Charbagh ( Lucknow  ) 🔝 8923113531 🔝  Cash PaymentTop Call Girls In Charbagh ( Lucknow  ) 🔝 8923113531 🔝  Cash Payment
Top Call Girls In Charbagh ( Lucknow ) 🔝 8923113531 🔝 Cash Paymentanilsa9823
 
Dubai Call Girls O528786472 Diabolic Call Girls In Dubai
Dubai Call Girls O528786472 Diabolic Call Girls In DubaiDubai Call Girls O528786472 Diabolic Call Girls In Dubai
Dubai Call Girls O528786472 Diabolic Call Girls In Dubaihf8803863
 
Online Social Shopping Motivation: A Preliminary Study
Online Social Shopping Motivation: A Preliminary StudyOnline Social Shopping Motivation: A Preliminary Study
Online Social Shopping Motivation: A Preliminary StudyAJHSSR Journal
 
CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...
CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...
CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...anilsa9823
 
Night 7k Call Girls Noida Sector 121 Call Me: 8448380779
Night 7k Call Girls Noida Sector 121 Call Me: 8448380779Night 7k Call Girls Noida Sector 121 Call Me: 8448380779
Night 7k Call Girls Noida Sector 121 Call Me: 8448380779Delhi Call girls
 

Recently uploaded (20)

定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一
定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一
定制(ENU毕业证书)英国爱丁堡龙比亚大学毕业证成绩单原版一比一
 
FULL ENJOY Call Girls In Mohammadpur (Delhi) Call Us 9953056974
FULL ENJOY Call Girls In Mohammadpur  (Delhi) Call Us 9953056974FULL ENJOY Call Girls In Mohammadpur  (Delhi) Call Us 9953056974
FULL ENJOY Call Girls In Mohammadpur (Delhi) Call Us 9953056974
 
c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...
c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...
c Starting with 5000/- for Savita Escorts Service 👩🏽‍❤️‍💋‍👨🏿 8923113531 ♢ Boo...
 
Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...
Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...
Top Astrologer, Kala ilam specialist in USA and Bangali Amil baba in Saudi Ar...
 
CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service 👖
CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service  👖CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service  👖
CALL ON ➥8923113531 🔝Call Girls Takrohi Lucknow best Female service 👖
 
🔝9953056974 🔝Call Girls In Mehrauli Escort Service Delhi NCR
🔝9953056974 🔝Call Girls In Mehrauli  Escort Service Delhi NCR🔝9953056974 🔝Call Girls In Mehrauli  Escort Service Delhi NCR
🔝9953056974 🔝Call Girls In Mehrauli Escort Service Delhi NCR
 
O9654467111 Call Girls In Dwarka Women Seeking Men
O9654467111 Call Girls In Dwarka Women Seeking MenO9654467111 Call Girls In Dwarka Women Seeking Men
O9654467111 Call Girls In Dwarka Women Seeking Men
 
Learn About the Rise of Instagram Pro in 2024
Learn About the Rise of Instagram Pro in 2024Learn About the Rise of Instagram Pro in 2024
Learn About the Rise of Instagram Pro in 2024
 
Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779
Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779
Night 7k Call Girls Noida New Ashok Nagar Escorts Call Me: 8448380779
 
Spotify AI DJ Deck - The Agency at University of Florida
Spotify AI DJ Deck - The Agency at University of FloridaSpotify AI DJ Deck - The Agency at University of Florida
Spotify AI DJ Deck - The Agency at University of Florida
 
9953056974 Young Call Girls In Kirti Nagar Indian Quality Escort service
9953056974 Young Call Girls In  Kirti Nagar Indian Quality Escort service9953056974 Young Call Girls In  Kirti Nagar Indian Quality Escort service
9953056974 Young Call Girls In Kirti Nagar Indian Quality Escort service
 
Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779
Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779
Night 7k Call Girls Pari Chowk Escorts Call Me: 8448380779
 
Call Girls In Patel Nagar Delhi 9654467111 Escorts Service
Call Girls In Patel Nagar Delhi 9654467111 Escorts ServiceCall Girls In Patel Nagar Delhi 9654467111 Escorts Service
Call Girls In Patel Nagar Delhi 9654467111 Escorts Service
 
Night 7k Call Girls Noida Sector 120 Call Me: 8448380779
Night 7k Call Girls Noida Sector 120 Call Me: 8448380779Night 7k Call Girls Noida Sector 120 Call Me: 8448380779
Night 7k Call Girls Noida Sector 120 Call Me: 8448380779
 
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Masudpur
Delhi  99530 vip 56974  Genuine Escort Service Call Girls in MasudpurDelhi  99530 vip 56974  Genuine Escort Service Call Girls in Masudpur
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Masudpur
 
Top Call Girls In Charbagh ( Lucknow ) 🔝 8923113531 🔝 Cash Payment
Top Call Girls In Charbagh ( Lucknow  ) 🔝 8923113531 🔝  Cash PaymentTop Call Girls In Charbagh ( Lucknow  ) 🔝 8923113531 🔝  Cash Payment
Top Call Girls In Charbagh ( Lucknow ) 🔝 8923113531 🔝 Cash Payment
 
Dubai Call Girls O528786472 Diabolic Call Girls In Dubai
Dubai Call Girls O528786472 Diabolic Call Girls In DubaiDubai Call Girls O528786472 Diabolic Call Girls In Dubai
Dubai Call Girls O528786472 Diabolic Call Girls In Dubai
 
Online Social Shopping Motivation: A Preliminary Study
Online Social Shopping Motivation: A Preliminary StudyOnline Social Shopping Motivation: A Preliminary Study
Online Social Shopping Motivation: A Preliminary Study
 
CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...
CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...
CALL ON ➥8923113531 🔝Call Girls Ashiyana Colony Lucknow best sexual service O...
 
Night 7k Call Girls Noida Sector 121 Call Me: 8448380779
Night 7k Call Girls Noida Sector 121 Call Me: 8448380779Night 7k Call Girls Noida Sector 121 Call Me: 8448380779
Night 7k Call Girls Noida Sector 121 Call Me: 8448380779
 

Build a Streaming Architecture with Kafka

  • 2. AGENDA 1/6/2021 © 2012 Telenav, Proprietary and Confidential 2 • Story: Build Netflix • Streaming tool: Kafka • What we could do?
  • 3. BUILD NETFLIX 1/6/2021 © 2012 Telenav, Proprietary and Confidential 3 Problems of grey boxes 1. Multiple techniques for arrow between different components 2. Dependency between two components
  • 4. MESSAGE TECHNOLOGY 1/6/2021 © 2012 Telenav, Proprietary and Confidential 4 Remove unnecessary dependency and complexity
  • 5. MORE REQUIREMENTS FOR THUMBNAIL 1/6/2021 © 2012 Telenav, Proprietary and Confidential 5
  • 6. MORE ADJUSTMENT FOR THUMBNAIL 1/6/2021 © 2012 Telenav, Proprietary and Confidential 6
  • 7. FINAL 1/6/2021 © 2012 Telenav, Proprietary and Confidential 7
  • 8. FINAL 1/6/2021 © 2012 Telenav, Proprietary and Confidential 8 • Universal microservices architecture • Maintaining independence of microservices by using lightweight communications between services via a uniform message-passing technology (such as Apache Kafka or MapR Streams) that is durable and high performance • Use of distributed files, NoSQL databases, and snapshot
  • 9. WHAT IS KAFKA 1/6/2021 © 2012 Telenav, Proprietary and Confidential 9
  • 10. WHY KAFKA • From Jay Kreps #1. Transporting data between system #2. Core of Analytics system • Original problem raised by Jay Kreps: When you use database to analysis data, during analysis time new data come in, how to reflect changes? Tails+Tails + Tails? MessageMQ? 1/6/2021 © 2012 Telenav, Proprietary and Confidential 10
  • 11. WHY KAFKA • Message durability(commit log, append only behavior) • Keep message sequence • Designed for large volume of data • One message be consumed by multiple consumer for different purpose • Message be consumed Just once for each topic 1/6/2021 © 2012 Telenav, Proprietary and Confidential 11
  • 12. KAFKA TERMINOLOGY • Message Broker: serve to persist data and handle request from both producer and consumer • Producer client: produce publish • Consumer client: consume subscribe • Topic: partition at message broker layer when have large and large volume of data could scale up, partition topic partition: concept of topic, parameters user specific or generated based on certain pattern 1/6/2021 © 2012 Telenav, Proprietary and Confidential 12
  • 13. KAFKA SCENARIOS • #1. online data to offline data Online data like page view/impression, -> offline hadoop system, kafka be consumed by ETL • #2. online data to online pipeline Emails/news feed • #3. Offline data to online pipeline Data be generated in offline data wareehouse -> online machine learning 1/6/2021 © 2012 Telenav, Proprietary and Confidential 13
  • 14. CORE OF KAFKA – REPLICATED DISTRIBUTE LOG Log is an append-only, totally-ordered sequence of records ordered by time 1/6/2021 © 2012 Telenav, Proprietary and Confidential 14
  • 15. WHY LOG IS IMPORTANT • The answer is that logs have a specific purpose: they record what happened and when • The two problems a log solves—ordering changes and distributing data—are even more important in distributed data systems. • One of the beautiful things about this approach is that the time stamps that index the log now act as the clock for the state of the replicas—you can describe each replica by a single number, the timestamp for the maximum log entry it has processed. This timestamp combined with the log uniquely captures the entire state of the replica. 1/6/2021 © 2012 Telenav, Proprietary and Confidential 15
  • 16. STATE MACHINE REPLICATION • State-Machine-> active active -> The active-active approach might log out the transformations to apply, say "+1", "*2", etc. Each replica would apply these transformations and hence go through the same set of values. • Primary-Backup -> Active-Passive -> a single master execute the transformations and log out the result, say "1", "3", "6", etc. 1/6/2021 © 2012 Telenav, Proprietary and Confidential 16
  • 17. 1/6/2021 © 2012 Telenav, Proprietary and Confidential 17
  • 18. RELATION WITH TELENAV • Decouples Data producer with consumer • Log Replay(A/B test, parallel test) 1/6/2021 © 2012 Telenav, Proprietary and Confidential 18 User Prob Data User Service Requset Data update Message Bus(Kafka) Live data(weather, traffic) Data analytics A/B test Live traffic generation Data preprocessing Kafka Logs data data data data
  • 19. REFERENCE • Kafka Stream Examples https://github.com/CodeBear801/kafka-streams- examples • Jay Kreps https://engineering.linkedin.com/distributed- systems/log-what-every-software-engineer-should- know-about-real-time-datas-unifying https://www.confluent.io/blog/stream-data-platform-1 1/6/2021 © 2012 Telenav, Proprietary and Confidential 19

Editor's Notes

  1. How Kafka improve transportation between systems Traditional solution: buffer data in memory, consumer could get it quickly Problem: lack of durability -> persistent on disk to achieve low latency and high throughput   Richer processing: each message could take different time