Graphql on production with AWS

by Bogdan Nedelcu

Slide 1

Slide 1 text

@bogdanned github.com/bogdanned GraphQL @production on AWS !1

Slide 2

Slide 2 text

@bogdanned github.com/bogdanned – The Wikipedia “In the context of computer programming, magic … it is used to describe code that handles complex tasks while hiding that complexity to present a simple interface. ” !2

Slide 3

Slide 3 text

@bogdanned github.com/bogdanned @production !3 - critical user impact - limited replicability - resilience && scalability$

Slide 4

Slide 4 text

@bogdanned github.com/bogdanned !4 - Schema Design @production - System Design with Graphql

Slide 5

Slide 5 text

@bogdanned github.com/bogdanned !5 - Schema Design @production - System Design with Graphql

Slide 6

Slide 6 text

@bogdanned github.com/bogdanned !6 - Superﬂuous Resolver Calls

Slide 7

Slide 7 text

@bogdanned github.com/bogdanned !7

Slide 8

Slide 8 text

@bogdanned github.com/bogdanned !8

Slide 9

Slide 9 text

@bogdanned github.com/bogdanned !9 So should we dataload the whole thing? Requests Saved = NP - NC

Slide 10

Slide 10 text

@bogdanned github.com/bogdanned !10 - Schema Design @production - System Design with Graphql

Slide 11

Slide 11 text

@bogdanned github.com/bogdanned !11 - Schema Design @production - System Design with Graphql

Slide 12

Slide 12 text

@bogdanned github.com/bogdanned What does GraphQL implies for our systems? !12

Slide 13

Slide 13 text

@bogdanned github.com/bogdanned !13 –The GraphQL Website “GraphQL is a query language for your API, and a server-side runtime for executing queries by using a type system you deﬁne for your data.”

Slide 14

Slide 14 text

@bogdanned github.com/bogdanned –Johnny Appleseed Under the hood … !14 - The Api Gateway pattern - The Enterprise Data Model pattern - The Integration Database pattern - The Backend for Front End pattern: BFFF

Slide 15

Slide 15 text

@bogdanned github.com/bogdanned –Johnny Appleseed Under the hood … !15 - The Api Gateway pattern - The Enterprise Data Model pattern - The Integration Database pattern - The Backend for Front End pattern: BFFF

Slide 16

Slide 16 text

@bogdanned github.com/bogdanned !16 - service client coupling - too many round trips - redundant edge functions

Slide 17

Slide 17 text

@bogdanned github.com/bogdanned !17

Slide 18

Slide 18 text

@bogdanned github.com/bogdanned !18 - reverse proxy - gateway ofﬂoading - request aggregation

Slide 19

Slide 19 text

@bogdanned github.com/bogdanned - single point of failure - vertical scaling - horizontal scaling ) - release coupling !19

Slide 20

Slide 20 text

@bogdanned github.com/bogdanned –Johnny Appleseed Under the hood … !20 - The Api Gateway pattern - The Enterprise Data Model pattern - The Integration Database pattern - The Backend for Front End pattern: BFFF

Slide 21

Slide 21 text

@bogdanned github.com/bogdanned !21 Apollo Federation -separation of concerns -distributed graph -development tooling

Slide 22

Slide 22 text

@bogdanned github.com/bogdanned !22 -heterogeneity -schema matching -schema mapping Federated Database

Slide 23

Slide 23 text

@bogdanned github.com/bogdanned – Phil Calcado, Some thoughts on GraphQL vs. BFF 2019 “ .. it is hard for me to believe that you can combine the needs of many different applications, owned by different teams, with different users and use cases, in a single schema” !23

Slide 24

Slide 24 text

@bogdanned github.com/bogdanned !24 Which amount?

Slide 25

Slide 25 text

@bogdanned github.com/bogdanned –Johnny Appleseed Under the hood … !25 - The Api Gateway pattern - The Enterprise Data Model pattern - The Integration Database pattern - The Backend for Front End pattern: BFFF ,