Never again offline?!? Experiences on the outstanding role of data in a large-scale mobile app ecosystem

Transcription

1 Never again offline?!? Experiences on the outstanding role of data in a large-scale mobile app ecosystem Dr. Matthias Naab, Fraunhofer IESE 1 Dr. Ralf Carbon, John Deere ETIC Susanne Braun, Fraunhofer IESE April 29, 2015 SATURN 2015, Baltimore, USA

2 2

3 3

4 John Deere Ecosystem My John Deere 4

5 Introducing Mobile to the John Deere Ecosystem My John Deere One app vs. many apps Native vs. hybrid or web Online vs. Offline Proprietary vs. standard data 5

6 Collaboration Project Duration: since Q Several phases and scope extensions Team ~ 6 full time persons Approach UX / Arch / Dev / QA 2 week iterations Testing with real end customers since summer 2014 Continuously feeding concepts into production version Re-newing the code base after ~6 months 6

7 Where to find Data in a Typical Architecture Document? 7

8 Where is Data Really??? 8

9 What does this mean for architecting? 9

10 We need an App Architecting it should be Simple! 1 10

11 We need an App Architecting it should be Simple! 1 JobManager ios App ios Data Modeling UI CoreData MVC DB: SQLite 11

12 Multiple Users Work on Shared Enterprise Data 2 12

13 Multiple Users Work on Shared Enterprise Data 2 Device 1: ipad JobManager ios App Device 2: iphone MyJobs ios App Cloud Backend Mobile Services HTTP REST JSON Logic Hibernate Data exchange via backend DB: MySQL 13

14 Apps Work on Shared Data 3 14

15 Apps Work on Shared Data 3 JobManager ios App UI CoreData? MyResources ios App UI CoreData DB: SQLite DB: SQLite ios <=7 : No shared data 15

16 Apps Work on Shared Data 3 JobManager ios App MyResources ios App Cloud Backend Mobile Services HTTP REST JSON Logic Data exchange via backend On app switch, update data Hibernate DB: MySQL 16

17 Users Have to Work Offline 4 17

18 Users Have to Work Offline 4 Device 1: ipad JobManager ios App UI Sync Engine CoreData Device 2: iphone MyJobs ios App UI Sync Engine CoreData job { "id": UUID", "revnr": "42, "syncstate": UPDATED" } Custom sync mechanism Sync service and engine UUID + RevisionID + SyncState Conflict detection + resolution Treatment of deletion High effort! DB: SQLite Sync Cloud Backend DB: SQLite Mobile Services Logic Hibernate DB: MySQL 18

19 Data has Different Characteristics 5 Master Data Field Transactional Data Job Entities Machine Assignment Person Value Objects GPS Coordinate GPS Coordinate Data classification Data modeling rules (e.g. transactional accesses master, usage of aggregates) Life-cycle of objects (deletion) 19

20 JD Operates Multiple Tenants on the Backend 6 Tenant 1 Tenant 2 Tenant n job { "id": UUID", tenantid": 1", UUID", "revnr": "42, "syncstate": UPDATED" } ipad ipad ipad ipad ipad ipad ipad ipad ipad ipad ipad ipad Cloud Backend DB: MySQL Separation of tenants with tenantid for each object General data provided by JD Sync has to respect tenants Cost-efficient operation Tenant 1 Data Tenant 2 Data General Data Tenant n Data 20

21 Multiple Users Share one Device 7 ios apps and devices not prepared for multi-user 21

22 Multiple Users Share one Device 7 Device 1: ipad JobManager ios App UI Sync Engine CoreData DB: SQLite Login mechanism for users Custom realization of SSO Usage of keychain group access for SSO tokens Sharing among apps via backend Removing data of logged out users from current SQLite DB User 1 Data User 2 Data 22

23 Connecting to the World 8 Machine Identification Device 1: ipad ios App Map Service Machine Data Cloud Backend Collecting data on the field Streaming map data Reusing weather data across tenants Sharing master data with other management systems Logic DB: MySQL Farm Management Weather Service 23

24 Increasing Amount of Data 9 Device 1: ipad JobManager ios App UI Sync Engine CoreData > objects Continuously growing Offline requires data sync Slow initial sync (> 60s) Filter old transactional data Compress Prefill SQLite DB file Deeply optimize slow CoreData storage DB: SQLite Sync Cloud Backend Mobile Services Logic Hibernate DB: MySQL 24

25 Increasing Frequency of Changes of Data 10 Device 1: iphone Logistics ios App Near real-time data (GPS, ) ~1 change/s Live distribution to devices Visualization and analysis HFR Cloud Backend HFW Sync Mob. Serv. CQRS: Separate interface for high frequency data Event-Sourcing Cassandra for fast writes In-memory for fast reads In Memory Cassandra Logic Hibernate DB: MySQL 25

26 Upgrading Apps 11 Device 1: ipad JobManager ios App Device 2: ipad JobManager ios App v 1 v 2 SQLite (model v 1 ) SQLite (model v 2 ) Cloud Backend S v 1 + Mapper S v 2 Logic Support multiple app versions Maintain single DB model Create mapper Hibernate DB: MySQL (model v 2 ) 26

27 Multiple Languages to be Supported 12 Language resources for texts Device 1: ipad JobManager ios App UI Device 2: ipad JobManager ios App UI Translated master, reference, and transactional data Metrics / units English German Cloud Backend Sync Mobile Services Logic Where to transform units? In which units to persist? Tenant-spanning analyses Hibernate DB: MySQL 27

28 Summary: Overview of Data Representations Device 1: ipad JobManager ios App UI Sync Engine CoreData DB: SQLite CoreData Objects Relational Tables JSON DTOs Java Objects Cassandra Column F. HFR In Memory HFW Cassandra Cloud Backend Sync Mob. Serv. Logic Hibernate DB: MySQL Java DTOs Java Entities Relational Tables 28

29 Lessons Learned Impact on Quality Attributes UX and performance: high impact on data architecture Security: makes offline capability even more difficult Maintainability: nearly always adversely impacted by concepts for other quality attributes 29

30 Lessons Learned Technical Aspects Offline capability: not out of the box Offline capability: costly to develop CoreData: requires in-depth technical know-how 30

31 Lessons Learned Data Modeling There is not THE single data model Data modeling is highly intertwined with architecture design Data modeling according to OOD / DDD is not enough 31

32 Key Takeaways Data is often neglected in architecture Data is highly related to quality attributes Design and document decisions around data! Farming is innovative and strongly uses IT 32

33 33

34 Dr. Matthias Naab Phone: Dr. Ralf Carbon Phone:

35 Appendix: Legend of Diagrams Component Computational Node Connector App Data object Yellow: Introduced or changed here 35