AppSyncを使いこなすためのDynamoDB設計パターン

Transcript

1. Amazon Web Services Japan Tsukagoshi Keisuke AppSync
    DynamoDB  GraphQL
    NoSQL  
2. Who am I ? e gd B A @ Mobile / DevOps / Serverless / Microservices W SJ ba W J c S=
3. DynamoDB 
   
   
4. DynamoDB • 
     •   •    
5. 78
     • SPOF  • .6,!3 AZ#"   • +/46*!16-%)205(#" '3&$5/
6. D : 9?7/;@50>A924 • 3A8>!#Read!Write((#  "0>A 924-:6.3*)& 'BE9?7/;@'C • ReadCapacityUnitBRCUC!WriteCapacityUnitBWCUC =1A0)& (%$ ' • 
   #,@<+@ 
7. -.%(+,$  •  %(+,$ • ',&   '"%#!*,) 
   
8. DynamoDB
   
9. DynamoDB
    • Scan  "!"   • CQRS "!
   "  
10. Scan
    
11. DynamoDB1 • (7IUE93.57 •
    /RUR.',DynamoDB :MQ@UBPT2-$7* % /"HULR9 (7 #!63( 1HULR9 (7 1#/"68)1:MQ@UBPT - /HULR2 1 +14-( • GSI9(7 •  1?SUKRD=TFQ;TIG>C9 (7&.-O; THULR-ANUJ-$7><Q.2/7><Q90(7 &.#-$3(
12. Product Database ID Type Price Description Products ID Title Category Fiction Producer Director … Videos ID Artist Title Genre … Albums ID Author Title Fiction Category Date … Books { Product ID, Type, Price, Description, Author, Title, Fiction, Category, Date, … } ID AlbumID Title Duration … Tracks ActorID VideoID ActorVideo ID Name Age Gender Bio … Actors { Product ID, Type, Price, Description, Artist, Title, Genre, Tracks: [ { Title1, Duration1 },{ Title2, Duration2, } ] … } { Product ID, Type, Price, Description, Title, Category, Fiction, Producer, Director, Actors: [ { ActorID, Name, Age, Gender, ShortBio }, … ] } Normalization Aggregation DynamoDB
    
13. RDBMS  DynamoDB 
     m Mo a n a aD M R o SBR n M a y M
14. " " !  !""  ""
       "!
     "  RDBMS  NoSQL 
15. %.'+  • GetItem • Partation Key 
     !%) • PutItem • 1 !%) • Update • 1 !%) • Delete • 1 !%) • • a G • • a P Q G • • G BKS
    Query  Scan "(,.#*-  1 MB &.$ 
16. &3)1
    +!2' DynamoDB API
     • (3& #.2"3  (5),%$ !&-  partition key = value  • 4 Item  • (*/!,0"3  Item  )
17. (1+/ ,%0) • Scan$"(1+/$*- • Scan$Query / GetItem / BatchGet !(1+/" • '&. '&.$Partition Key$
     #"!" • Scan • (1+/*-"
18. ,>6:5?> #!Partition key • DB .>3> *)2?1>4';9?0ID +:? 7-2<>/? 
    ) • 2?1>4ID +0.=#  ) ! 8?5#!%(#&":?7-2<>/?+
     +$ Partition Key Attribute1 Attribute2 TransactionID OrderID Order_Date 1111111 Customer1-1 2016-05-17 01.36.45 1111112 Customer1-2 2016-05-18 01.36.30 1111113 Customer2-1 2016-05-18 01.36.30
19. CQRS 
    
20. CQRS • Command Query Responsibility Segregation • Command (write) % Query (read) '
    ,* • Why • =C41C4& #7C5<7?(* • 2B0?7C59C4 & CRUD "%* • Command / Query  ,*%$* • Benefit • + +&&# !7C5<7?,* • Read / Write , #41C?$* • 8;-BA36/%.:@C2>B, )+*
21. AWS Lambda DynamoDB Stream Query service 2 Amazon ES Query service 1 Query service 3 Amazon Redshift Amazon DynamoDB Command side Query side Amazon DynamoDB DynamoDB
    Event Sourcing
22. AWS Lambda Amazon DynamoDB Amazon API Gateway Amazon DynamoDB Stream AWS Lambda Amazon S3 Amazon ES Amazon Athena Amazon DynamoDB Query side Command side ü DynamoDB Streams  Lambda Event Sourcing ü QueryCommand
      Command Query Responsibility Segregation Serverless
    CQRS 
23. AppSync  CQRS &#*) Amazon DynamoDB ElasticSearch     Command  DynamoDBPut Query
    !*(*%' "*$*#
24. 
    :  
25. RDBMS    Rows Primary Key Index
     ID  1 2 3     SA SA SA Columns
26. DynamoDB  Data
    Partition Key Sort Key  SA ID ColA 1 1 1    Data
     SA ColA    Partition KeySort Key  ID 1 1 1 Attribute
27. 
     on S R y M R DS M S B yM n R y m n n S M a S