SlideShare a Scribd company logo
1 of 41
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
AWS re:INVENT
Airbnb Runs on Aurora
S i r i s h C h a n d r a s e k a r a n – P r i n c i p l e P r o d u c t M a n a g e r , A m a z o n A u r o r a
X i n y a o H u – E n g i n e e r M a n a g e r , A i r b n b
R u o b i n g L i – S o f t w a r e E n g i n e e r , A i r b n b
D A T 3 3 1
November 29, 2017
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Airbnb Overview
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Airbnb Overview
3,000,000
listings
65,000
cities
191
countries
An online marketplace and hospitality service
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Airbnb Architecture Overview
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Database Business Requirement
• Avg latency: 1-2ms
• P95 latency: 10ms
• Average DB queries on a listing detail page: ~30
• QPS: 100K for the biggest cluster
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Airbnb’s Database Deployment Overview
• Mostly MySQL with very few Postgres / Oracle.
• All MySQL deployed on RDS@AWS from 2011.
• Hundreds of RDS clusters deployed. One of the largest RDS deployments in AWS.
• Managed by terraform.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Challenges
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Weekly Traffic vs. Response Time
October 2013 to Present (Last 4 Years)
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Number of Services / Databases Running In Prod
Jan 2016 to Present (Past 2 years)
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Data Growth
Data Growth for one cluster from Feb 2017 to Aug 2017
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Operational Challenges when Running at Web-Scale
• Takes more than a day to create a replica for big clusters.
• Takes more than 16 hours to restore a replica for big clusters.
• Need to warm up the disk layer for RDS.
• Frequent jitter.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Operational Challenges when Running at Web-Scale
Jitter
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Operational Challenges when Running at Web-Scale
Jitter
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Operational Challenges when Running at Web-Scale
Jitter
10% Airbnb’s down-time can be
attributed to latency jitter.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Alternatives
EC2 MySQL + DBA Team Aurora
• High Performance Local SSD.
• More controls and visibility to the system.
• Need to handle sharding in the near term.
• Need to staff a DBA team.
• No jitter
• Scale up to 64TB.
• Fully managed by AWS.
• More appealing features which are not
available in standard MySQL builds.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Evaluation Metrics
Performance Compatibility Failure Scenarios Operational Overhead
• Latency
• Throughput
• Read / write
headroom
Evaluating on real production workload and use cases
• Ruby / Java Clients
• Transaction
isolation
• Durability setting
• Monitoring
• Tooling support
• Disk failure
• Network blip
• Master failover
• Create cluster
• Create read replica
• Change parameter
group
• Point in time
recovery (PITR)
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
DBReplay
• In-house developed load testing framework.
• Replay production workload on restored production DB.
• Stress test DB with 1x ~ 10x workload.
• Help database headroom projection.
• Verify new infrastructure project (Aurora).
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
DBReplay
Overview
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
DBReplay
User Interface
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
DBReplay
MySQL / Aurora side by side comparison
Aurora
MySQL 5.7
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Performance Metrics Summary*
2x
Read head room
4x
Write head room
- 75 %
Write Latency
* based on a write heavy workload, result may vary for different workload
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Non-Performance Metrics Summary
Compatibility Ruby / Java Clients Fully compatible with Ruby clients and
JDBC
Compatibility Transaction isolation Same behavior as MySQL
Compatibility Durability settings Does not lose binlog in crash
Compatibility Monitoring Comparable CloudWatch and MySQL
metrics
Compatibility Tooling support Fully compatible with mysqlbinlog /
mydumper / terraform
Failure Scenarios Network blip none
Failure Scenarios Time to do a master failover 2 minutes -> 20 seconds
Operational Overhead Time to create a read replica 1 day -> 5 minutes
Operational Overhead PITR 10 hours -> 1 hour
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Jitter
0% No jitter was observed during
Aurora evaluation.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Requirements
• Minimize disruptions on end users and our product teams.
• Scalable enough to handle 100+ clusters.
• Migration should be revertible.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration Timeline
July 2017 Sep 2017 Oct 2017
Start Aurora
Evaluation.
Finish the
evaluation. Start
running replica
traffic on Aurora
Migrate a few of
the most important
masters to Aurora.
Still keep RDS
MySQL fleet. Can
revert the
migration if
necessary.
Failover all
masters while
keeping RDS
MySQL fleet.
Dec 2017 Jan 2018
Delete all RDS
MySQL instances.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Terraform
• Manages RDS resources with code.
• Automates DB cluster creation, update and deletion.
• Tracks complete history of DB changes.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration Stage 1
Create Aurora Replicas
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration Stage 2
Shift Slave Traffic from RDS MySQL to Aurora
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
DBProxy
• Maxscale open source MySQL database proxy.
• In-house developed features.
• Configurable connection pooling
• Requests throttling and back pressure
• Blacklist queries rejection
• Fast connections refresh
• Drop-in deployment as dbproxy services for core RDS databases.
• Allow scaling web service tier independently from database tier
• Proxy tier scale horizontally with capacity increase
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
DBProxy
Fast Connection Refresh
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
DBProxy
Fast Connection Refresh Cont’d
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration Stage 3
Master Promotion
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration Stage 3
Master Promotion cont’d
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration Stage 3
Master Promotion cont’d
• < 1 minute of write down time. No read down time.
• Stage 3 is revertible.
• Repeat this process for every cluster.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration Progress
25 %
Clusters migrated
70 %
Traffic migrated
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Conclusion & Future Work
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Benefits of Aurora
RDS MySQL Aurora
Read headroom 1x 2x
Write headroom 1x 4x
Jitter Yes No
Replica lag Up to minutes 20ms
Time to do a master
failover
2 minutes 20 seconds
Time to create replica 1 day 5 minutes
PITR 10 hours 1 hour
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Leverage Aurora features
• Clone
• Fast DDL
• Load data into S3 for disaster recovery
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Thank You!
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.

More Related Content

What's hot

AWSサービスで実現するEightの行動ログ活用基盤
AWSサービスで実現するEightの行動ログ活用基盤AWSサービスで実現するEightの行動ログ活用基盤
AWSサービスで実現するEightの行動ログ活用基盤Tetsuya Mase
 
마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)
마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)
마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)Amazon Web Services Korea
 
データ活用を加速するAWS分析サービスのご紹介
データ活用を加速するAWS分析サービスのご紹介データ活用を加速するAWS分析サービスのご紹介
データ活用を加速するAWS分析サービスのご紹介Amazon Web Services Japan
 
20210216 AWS Black Belt Online Seminar AWS Database Migration Service
20210216 AWS Black Belt Online Seminar AWS Database Migration Service20210216 AWS Black Belt Online Seminar AWS Database Migration Service
20210216 AWS Black Belt Online Seminar AWS Database Migration ServiceAmazon Web Services Japan
 
20210316 AWS Black Belt Online Seminar AWS DataSync
20210316 AWS Black Belt Online Seminar AWS DataSync20210316 AWS Black Belt Online Seminar AWS DataSync
20210316 AWS Black Belt Online Seminar AWS DataSyncAmazon Web Services Japan
 
20190319 AWS Black Belt Online Seminar Amazon FSx for Windows Server
20190319 AWS Black Belt Online Seminar Amazon FSx for Windows Server20190319 AWS Black Belt Online Seminar Amazon FSx for Windows Server
20190319 AWS Black Belt Online Seminar Amazon FSx for Windows ServerAmazon Web Services Japan
 
20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...
20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...
20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...Amazon Web Services Japan
 
20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-
20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-
20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-Amazon Web Services Japan
 
20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...
20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...
20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...Amazon Web Services Japan
 
From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...Amazon Web Services Korea
 
AWS Black Belt Techシリーズ Elastic Load Balancing (ELB)
AWS Black Belt Techシリーズ  Elastic Load Balancing (ELB)AWS Black Belt Techシリーズ  Elastic Load Balancing (ELB)
AWS Black Belt Techシリーズ Elastic Load Balancing (ELB)Amazon Web Services Japan
 
Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...
Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...
Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...Amazon Web Services
 
サーバーレスアーキテクチャのすすめ(公開版)
サーバーレスアーキテクチャのすすめ(公開版)サーバーレスアーキテクチャのすすめ(公開版)
サーバーレスアーキテクチャのすすめ(公開版)Keisuke Kadoyama
 
AWS Black Belt Online Seminar 2018 AWS Well-Architected Framework
AWS Black Belt Online Seminar 2018 AWS Well-Architected FrameworkAWS Black Belt Online Seminar 2018 AWS Well-Architected Framework
AWS Black Belt Online Seminar 2018 AWS Well-Architected FrameworkAmazon Web Services Japan
 
High Performance Computing (HPC) on AWS 101
High Performance Computing (HPC) on AWS 101High Performance Computing (HPC) on AWS 101
High Performance Computing (HPC) on AWS 101Amazon Web Services
 
AWS EMR Cost optimization
AWS EMR Cost optimizationAWS EMR Cost optimization
AWS EMR Cost optimizationSANG WON PARK
 

What's hot (20)

AWSサービスで実現するEightの行動ログ活用基盤
AWSサービスで実現するEightの行動ログ活用基盤AWSサービスで実現するEightの行動ログ活用基盤
AWSサービスで実現するEightの行動ログ活用基盤
 
마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)
마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)
마이크로 서비스를 위한 AWS Cloud Map & App Mesh - Saeho Kim (AWS Solutions Architect)
 
データ活用を加速するAWS分析サービスのご紹介
データ活用を加速するAWS分析サービスのご紹介データ活用を加速するAWS分析サービスのご紹介
データ活用を加速するAWS分析サービスのご紹介
 
20210216 AWS Black Belt Online Seminar AWS Database Migration Service
20210216 AWS Black Belt Online Seminar AWS Database Migration Service20210216 AWS Black Belt Online Seminar AWS Database Migration Service
20210216 AWS Black Belt Online Seminar AWS Database Migration Service
 
20210316 AWS Black Belt Online Seminar AWS DataSync
20210316 AWS Black Belt Online Seminar AWS DataSync20210316 AWS Black Belt Online Seminar AWS DataSync
20210316 AWS Black Belt Online Seminar AWS DataSync
 
20190319 AWS Black Belt Online Seminar Amazon FSx for Windows Server
20190319 AWS Black Belt Online Seminar Amazon FSx for Windows Server20190319 AWS Black Belt Online Seminar Amazon FSx for Windows Server
20190319 AWS Black Belt Online Seminar Amazon FSx for Windows Server
 
20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...
20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...
20180704(20190520 Renewed) AWS Black Belt Online Seminar Amazon Elastic File ...
 
AWS Database Migration Service ご紹介
AWS Database Migration Service ご紹介AWS Database Migration Service ご紹介
AWS Database Migration Service ご紹介
 
20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-
20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-
20210330 AWS Black Belt Online Seminar AWS Glue -Glue Studioを使ったデータ変換のベストプラクティス-
 
20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...
20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...
20191002 AWS Black Belt Online Seminar Amazon EC2 Auto Scaling and AWS Auto S...
 
From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...
 
AWS Black Belt Techシリーズ Elastic Load Balancing (ELB)
AWS Black Belt Techシリーズ  Elastic Load Balancing (ELB)AWS Black Belt Techシリーズ  Elastic Load Balancing (ELB)
AWS Black Belt Techシリーズ Elastic Load Balancing (ELB)
 
Black Belt Online Seminar Amazon CloudWatch
Black Belt Online Seminar Amazon CloudWatchBlack Belt Online Seminar Amazon CloudWatch
Black Belt Online Seminar Amazon CloudWatch
 
Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...
Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...
Amazon DynamoDB Under the Hood: How We Built a Hyper-Scale Database (DAT321) ...
 
Netflix and Open Source
Netflix and Open SourceNetflix and Open Source
Netflix and Open Source
 
サーバーレスアーキテクチャのすすめ(公開版)
サーバーレスアーキテクチャのすすめ(公開版)サーバーレスアーキテクチャのすすめ(公開版)
サーバーレスアーキテクチャのすすめ(公開版)
 
AWS Black Belt Online Seminar 2018 AWS Well-Architected Framework
AWS Black Belt Online Seminar 2018 AWS Well-Architected FrameworkAWS Black Belt Online Seminar 2018 AWS Well-Architected Framework
AWS Black Belt Online Seminar 2018 AWS Well-Architected Framework
 
AWS Black Belt online seminar 2017 Snowball
AWS Black Belt online seminar 2017 SnowballAWS Black Belt online seminar 2017 Snowball
AWS Black Belt online seminar 2017 Snowball
 
High Performance Computing (HPC) on AWS 101
High Performance Computing (HPC) on AWS 101High Performance Computing (HPC) on AWS 101
High Performance Computing (HPC) on AWS 101
 
AWS EMR Cost optimization
AWS EMR Cost optimizationAWS EMR Cost optimization
AWS EMR Cost optimization
 

Similar to Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017

Scale Website dan Mobile Applications Anda di AWS hingga 10 juta pengguna
Scale Website dan Mobile Applications Anda di AWS hingga 10 juta penggunaScale Website dan Mobile Applications Anda di AWS hingga 10 juta pengguna
Scale Website dan Mobile Applications Anda di AWS hingga 10 juta penggunaAmazon Web Services
 
Scaling Up to Your First 10 Million Users
Scaling Up to Your First 10 Million UsersScaling Up to Your First 10 Million Users
Scaling Up to Your First 10 Million UsersAmazon Web Services
 
How to Design a Multi-Region Active-Active Architecture
How to Design a Multi-Region Active-Active ArchitectureHow to Design a Multi-Region Active-Active Architecture
How to Design a Multi-Region Active-Active ArchitectureAmazon Web Services
 
Journey Towards Scaling Your API to 10 Million Users
Journey Towards Scaling Your API to 10 Million UsersJourney Towards Scaling Your API to 10 Million Users
Journey Towards Scaling Your API to 10 Million UsersAdrian Hornsby
 
Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...
Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...
Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...Amazon Web Services
 
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017Amazon Web Services
 
ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...
ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...
ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...Amazon Web Services
 
ATC301-Big Data & Analytics for Manufacturing Operations
ATC301-Big Data & Analytics for Manufacturing OperationsATC301-Big Data & Analytics for Manufacturing Operations
ATC301-Big Data & Analytics for Manufacturing OperationsAmazon Web Services
 
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...Amazon Web Services
 
ARC201_Scaling Up to Your First 10 Million Users
ARC201_Scaling Up to Your First 10 Million UsersARC201_Scaling Up to Your First 10 Million Users
ARC201_Scaling Up to Your First 10 Million UsersAmazon Web Services
 
Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?Amazon Web Services
 
ARC319_Multi-Region Active-Active Architecture
ARC319_Multi-Region Active-Active ArchitectureARC319_Multi-Region Active-Active Architecture
ARC319_Multi-Region Active-Active ArchitectureAmazon Web Services
 
How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...
How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...
How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...Amazon Web Services
 
Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...
Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...
Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...Amazon Web Services
 
DAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the CloudDAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the CloudAmazon Web Services
 
Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017
Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017
Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017Amazon Web Services
 
WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...
WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...
WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...Amazon Web Services
 
Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...
Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...
Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...Amazon Web Services
 
A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...
A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...
A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...Amazon Web Services
 
Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...
Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...
Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...Amazon Web Services
 

Similar to Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017 (20)

Scale Website dan Mobile Applications Anda di AWS hingga 10 juta pengguna
Scale Website dan Mobile Applications Anda di AWS hingga 10 juta penggunaScale Website dan Mobile Applications Anda di AWS hingga 10 juta pengguna
Scale Website dan Mobile Applications Anda di AWS hingga 10 juta pengguna
 
Scaling Up to Your First 10 Million Users
Scaling Up to Your First 10 Million UsersScaling Up to Your First 10 Million Users
Scaling Up to Your First 10 Million Users
 
How to Design a Multi-Region Active-Active Architecture
How to Design a Multi-Region Active-Active ArchitectureHow to Design a Multi-Region Active-Active Architecture
How to Design a Multi-Region Active-Active Architecture
 
Journey Towards Scaling Your API to 10 Million Users
Journey Towards Scaling Your API to 10 Million UsersJourney Towards Scaling Your API to 10 Million Users
Journey Towards Scaling Your API to 10 Million Users
 
Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...
Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...
Case Study: Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment - DAT33...
 
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
 
ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...
ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...
ARC306_High Resiliency & Availability Of Online Entertainment Communities Usi...
 
ATC301-Big Data & Analytics for Manufacturing Operations
ATC301-Big Data & Analytics for Manufacturing OperationsATC301-Big Data & Analytics for Manufacturing Operations
ATC301-Big Data & Analytics for Manufacturing Operations
 
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
 
ARC201_Scaling Up to Your First 10 Million Users
ARC201_Scaling Up to Your First 10 Million UsersARC201_Scaling Up to Your First 10 Million Users
ARC201_Scaling Up to Your First 10 Million Users
 
Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?
 
ARC319_Multi-Region Active-Active Architecture
ARC319_Multi-Region Active-Active ArchitectureARC319_Multi-Region Active-Active Architecture
ARC319_Multi-Region Active-Active Architecture
 
How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...
How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...
How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Event...
 
Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...
Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...
Case Study: Ola Cabs Uses Amazon EBS and Elastic Volumes to Maximize MySQL De...
 
DAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the CloudDAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the Cloud
 
Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017
Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017
Running Oracle Databases on Amazon RDS - DAT313 - re:Invent 2017
 
WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...
WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...
WIN301-Migrating Microsoft SQL Server Databases to AWS-Best Practices and Pat...
 
Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...
Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...
Migrating Microsoft SQL Server Databases to AWS – Best Practices and Patterns...
 
A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...
A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...
A Practitioner’s Guide on Migrating to, and Running on Amazon Aurora - DAT315...
 
Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...
Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...
Tinder and DynamoDB: It's a Match! Massive Data Migration, Zero Down Time - D...
 

More from Amazon Web Services

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Amazon Web Services
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Amazon Web Services
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateAmazon Web Services
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSAmazon Web Services
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Amazon Web Services
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Amazon Web Services
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...Amazon Web Services
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsAmazon Web Services
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareAmazon Web Services
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSAmazon Web Services
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAmazon Web Services
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareAmazon Web Services
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWSAmazon Web Services
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckAmazon Web Services
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without serversAmazon Web Services
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...Amazon Web Services
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceAmazon Web Services
 

More from Amazon Web Services (20)

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS Fargate
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWS
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot
 
Open banking as a service
Open banking as a serviceOpen banking as a service
Open banking as a service
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
 
Computer Vision con AWS
Computer Vision con AWSComputer Vision con AWS
Computer Vision con AWS
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatare
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e web
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWS
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch Deck
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without servers
 
Fundraising Essentials
Fundraising EssentialsFundraising Essentials
Fundraising Essentials
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container Service
 

Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017

  • 1. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AWS re:INVENT Airbnb Runs on Aurora S i r i s h C h a n d r a s e k a r a n – P r i n c i p l e P r o d u c t M a n a g e r , A m a z o n A u r o r a X i n y a o H u – E n g i n e e r M a n a g e r , A i r b n b R u o b i n g L i – S o f t w a r e E n g i n e e r , A i r b n b D A T 3 3 1 November 29, 2017
  • 2. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Airbnb Overview
  • 3. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Airbnb Overview 3,000,000 listings 65,000 cities 191 countries An online marketplace and hospitality service
  • 4. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Airbnb Architecture Overview
  • 5. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Database Business Requirement • Avg latency: 1-2ms • P95 latency: 10ms • Average DB queries on a listing detail page: ~30 • QPS: 100K for the biggest cluster
  • 6. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Airbnb’s Database Deployment Overview • Mostly MySQL with very few Postgres / Oracle. • All MySQL deployed on RDS@AWS from 2011. • Hundreds of RDS clusters deployed. One of the largest RDS deployments in AWS. • Managed by terraform.
  • 7. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Challenges
  • 8. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Weekly Traffic vs. Response Time October 2013 to Present (Last 4 Years)
  • 9. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Number of Services / Databases Running In Prod Jan 2016 to Present (Past 2 years)
  • 10. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Data Growth Data Growth for one cluster from Feb 2017 to Aug 2017
  • 11. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Operational Challenges when Running at Web-Scale • Takes more than a day to create a replica for big clusters. • Takes more than 16 hours to restore a replica for big clusters. • Need to warm up the disk layer for RDS. • Frequent jitter.
  • 12. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Operational Challenges when Running at Web-Scale Jitter
  • 13. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Operational Challenges when Running at Web-Scale Jitter
  • 14. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Operational Challenges when Running at Web-Scale Jitter 10% Airbnb’s down-time can be attributed to latency jitter.
  • 15. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Alternatives EC2 MySQL + DBA Team Aurora • High Performance Local SSD. • More controls and visibility to the system. • Need to handle sharding in the near term. • Need to staff a DBA team. • No jitter • Scale up to 64TB. • Fully managed by AWS. • More appealing features which are not available in standard MySQL builds.
  • 16. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Evaluation Metrics Performance Compatibility Failure Scenarios Operational Overhead • Latency • Throughput • Read / write headroom Evaluating on real production workload and use cases • Ruby / Java Clients • Transaction isolation • Durability setting • Monitoring • Tooling support • Disk failure • Network blip • Master failover • Create cluster • Create read replica • Change parameter group • Point in time recovery (PITR)
  • 17. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. DBReplay • In-house developed load testing framework. • Replay production workload on restored production DB. • Stress test DB with 1x ~ 10x workload. • Help database headroom projection. • Verify new infrastructure project (Aurora).
  • 18. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. DBReplay Overview
  • 19. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. DBReplay User Interface
  • 20. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. DBReplay MySQL / Aurora side by side comparison Aurora MySQL 5.7
  • 21. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Performance Metrics Summary* 2x Read head room 4x Write head room - 75 % Write Latency * based on a write heavy workload, result may vary for different workload
  • 22. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Non-Performance Metrics Summary Compatibility Ruby / Java Clients Fully compatible with Ruby clients and JDBC Compatibility Transaction isolation Same behavior as MySQL Compatibility Durability settings Does not lose binlog in crash Compatibility Monitoring Comparable CloudWatch and MySQL metrics Compatibility Tooling support Fully compatible with mysqlbinlog / mydumper / terraform Failure Scenarios Network blip none Failure Scenarios Time to do a master failover 2 minutes -> 20 seconds Operational Overhead Time to create a read replica 1 day -> 5 minutes Operational Overhead PITR 10 hours -> 1 hour
  • 23. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Jitter 0% No jitter was observed during Aurora evaluation.
  • 24. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration
  • 25. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Requirements • Minimize disruptions on end users and our product teams. • Scalable enough to handle 100+ clusters. • Migration should be revertible.
  • 26. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration Timeline July 2017 Sep 2017 Oct 2017 Start Aurora Evaluation. Finish the evaluation. Start running replica traffic on Aurora Migrate a few of the most important masters to Aurora. Still keep RDS MySQL fleet. Can revert the migration if necessary. Failover all masters while keeping RDS MySQL fleet. Dec 2017 Jan 2018 Delete all RDS MySQL instances.
  • 27. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Terraform • Manages RDS resources with code. • Automates DB cluster creation, update and deletion. • Tracks complete history of DB changes.
  • 28. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration Stage 1 Create Aurora Replicas
  • 29. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration Stage 2 Shift Slave Traffic from RDS MySQL to Aurora
  • 30. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. DBProxy • Maxscale open source MySQL database proxy. • In-house developed features. • Configurable connection pooling • Requests throttling and back pressure • Blacklist queries rejection • Fast connections refresh • Drop-in deployment as dbproxy services for core RDS databases. • Allow scaling web service tier independently from database tier • Proxy tier scale horizontally with capacity increase
  • 31. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. DBProxy Fast Connection Refresh
  • 32. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. DBProxy Fast Connection Refresh Cont’d
  • 33. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration Stage 3 Master Promotion
  • 34. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration Stage 3 Master Promotion cont’d
  • 35. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration Stage 3 Master Promotion cont’d • < 1 minute of write down time. No read down time. • Stage 3 is revertible. • Repeat this process for every cluster.
  • 36. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration Progress 25 % Clusters migrated 70 % Traffic migrated
  • 37. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Conclusion & Future Work
  • 38. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Benefits of Aurora RDS MySQL Aurora Read headroom 1x 2x Write headroom 1x 4x Jitter Yes No Replica lag Up to minutes 20ms Time to do a master failover 2 minutes 20 seconds Time to create replica 1 day 5 minutes PITR 10 hours 1 hour
  • 39. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Leverage Aurora features • Clone • Fast DDL • Load data into S3 for disaster recovery
  • 40. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Thank You!
  • 41. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.