Events, Microservices and Data Ben Stopford Office of the CTO, Confluent

A new category of infrastructure for data that moves Relational Database Data warehouse Messaging Complex Event Processing NoSQL & Big Data Event Streaming 1970 1980 1990 2000 2010 2020

No content

Event Storage Kafka stores petabytes of data Stream Processing Real-time processing over streams and tables Scalability Clusters of hundreds of machines. Global. + + + Roots in big data messaging

THREE TRENDS IN STREAMING Events become first-class citizens Microservices: ‘provision data’ as a service Stream Processing becomes a ‘database’ for the event-driven world

THREE TRENDS IN STREAMING Events become first-class citizens Microservices: ‘provision data’ as a service Stream Processing becomes a ‘database’ for the event-driven world

Event First: what changes? • The way we think about data • The way we architect • The way we evolve an organization • The way we program

What is an Event? A fact. A real-world incident, captured by a system. An payment A page view A log line A sensor reading

Events come in streams Apps M onitoring Security Apps Apps L A T F O R M Event Stream Order of events is important Apps M onitoring Apps Apps O R M Older Newer

Stream Processors Combine, Filter & Summarize Event Streams KSQL

Increasing Complexity Apps Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M Apps Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M App Apps Search NoSQL Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M App Apps Search NoSQL Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Mon Sec Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Apps App S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL S T R E A M I N Apps Search NoSQL Apps DWH S T R E A M I N G P L AT App Apps Search NoSQL Apps S T R E A M I N G P L AT Apps Search NoSQL Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL S T Apps Search NoSQL DWH S T R E App Apps Apps Search Apps App Apps Apps Apps Search Apps Apps App Apps Apps Search App Monolith Distributed Monolith Microservices Event-Driven Microservices Evolution of software systems

Increasing Complexity Apps Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M Apps Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M App Apps Search NoSQL Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M App Apps Search NoSQL Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Mon Sec Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Apps App S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL S T R E A M I N Apps Search NoSQL Apps DWH S T R E A M I N G P L AT App Apps Search NoSQL Apps S T R E A M I N G P L AT Apps Search NoSQL Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL S T Apps Search NoSQL DWH S T R E App Apps Apps Search Apps App Apps Apps Apps Search Apps Apps App Apps Apps Search App Kafka Evolution of software systems Monolith Distributed Monolith Microservices Event-Driven Microservices

Apps Search NoSQL Mo Se Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Monitoring Security Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Search NoSQL Apps Apps S T R E A M I N G P L AT F O R M Apps Search NoSQL Monitoring Security Apps Apps DWH Hadoop S T R E A M I N G P L AT F O R M App Apps Apps Apps Apps Apps Apps App Apps Apps Apps Apps Apps Apps App Apps Apps Apps Apps Apps Apps App Apps Apps Ap Apps Apps Apps App Apps Apps Apps Apps Apps App Apps Apps Apps Apps Apps Apps App Apps Apps Apps Apps Apps Apps App Backend services grow faster than the front end Tightly coupled Loosely coupled

Event shift the way we think Monolithic Approach -a database -a variable -a RPC -a user clicking buttons Event First Approach - an event stream - a “data” flow - a stream processor - software talking to software

Event-first changes how data is stored

Traditional systems use mutable state DB This isn’t wrong, it’s just lossy Apps Search Mon Apps Apps S T R E A M I N G P L A T F O R M

Events record the a timeseries of what actually happened. Shopping Cart Events 2 Trousers added 1 Jumper added 1 Trousers removed 1 Hat added Checkout Shopping Cart Event User Journey 12.42 12.44 12.49 12.50 12.59

Stored as a stream Stored statefully (think DB) 12.42 12.44 12.49 12.50 12.59 Information lost! Event User Journey

12.42 12.44 12.49 12.50 12.59 We can derive the current state (but not the other way around) Apps Apps DERIVE Stream Processor

Observe, Store, Project a form of “Event Sourcing” Familiar Stateful View LOSSY PROJECTION Stream = Exactly what happened

Events tell us not just what happened but also how it happened

The End State The sequencing of moves is often more important than the end state The Game

A stock price: observe the game, not just the current state

A customer journey: observe everything

Formula 1

Formula 1: Observe the game, optimize the end state now and in the future End state

Formulae 1 – High-Level Architecture • 400 Sensors on car • 70,000 derivative measures • Analyzed in real time • Tire modelling • Racing line • Aerodynamics • Machine Learning and Physics Models. • Replayed later for post race analysis. Race Track HQ e.g. Tire modelling: - Temp - Pressure - Suspension compression Stream Processing

Post race analysis

ML Source of Truth Retain events, rewind and replay the stream processor New derivative measures

Another form of “Event Sourcing” - Record what happened - Rewind, replay and rederive (View, App, ML, Physics Model etc.)

New York Times Source of Truth Every article since 1851 https://www.confluent.io/blog/publishing-apache-kafka-new-york-times/ Normalized assets (images, articles, bylines, tags all separate messages) Denormalized into “Content View”

Billing Shipping Fraud Fraud CUSTOMER ANALYTICS STORED EVENTS Stored Events used to create many different ‘views’

Event Streams Orders Payments Customers Distinct Visits Destination Spark Postgres Lambda Other Kafka Select Event Source Stream Processing SELECT * FROM ORDERS O, CUSTOMERS C WHERE O.REGION = ‘EU’ AND C.TYPE = ‘Platinum’ Msgs/Day Customers Stream Processing Postgres Lambda Orders History 1w All Event Streaming Platforms make data self service (via events)

THREE TRENDS IN STREAMING Events become first-class citizens Microservices: provision data as a service Stream Processing becomes a ‘database’ for the event-driven world

THREE TRENDS IN STREAMING Events become first-class citizens Microservices: provision data as a service Stream Processing becomes a ‘database’ for the event-driven world

A future of Aside: Serverless Functions

Bad Analogy Serverless Functions are like very small microservices

Serverless Functions (FaaS) • Write a function • Upload • Configure a trigger (HTTP, Messaging, Object Store, Database, Timer etc.) Request Respond Event Source

FaaS in a Nutshell • Fully managed • Pay for execution time • Auto-scales with load • Short lived

Better Analogy Serverless Functions small event-driven containers

FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS More events created -> More “containers” are spun up in response More images stored More functions are triggered to create thumbnails

In the future it seems unlikely that infrastructure will respond to events. But where will we manage our data?

Mr Microservices

Event-Streaming approaches this from a different angle

FaaS is event-driven But it isn’t event-streaming

Customers Event Source Orders Event Source Payments Event Source Serverless Way: event driven but not streaming FaaS/μS FaaS/μS FaaS/μS Use Case: Email a customer the details of their order

Payments Event Source Increase Load: 100 concurrent functions doing IO. FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS

Payments Event Source What if we’re only interested in platinum clients? -> 100 functions, 100 DB calls, 2 emails. FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS FaaS/μS

FaaS FaaS FaaS KSQL KSQL in-lines data processing into the flow FaaS FaaS Stateless “Business Logic” Filters what isn’t needed Orders Payments Customers Everything now enriched, summarized in the events

FaaS Traditional Application Event-Driven Application Application Database KSQL Stateful “Database” FaaS FaaS FaaS FaaS FaaS Streaming Stateless Stateless Stateless Compute Layer Stream processors acting as the ‘database’ of the event driven world.

THREE TRENDS • Events as first-class citizens • Microservices: provision data as a service • Stream Processing becomes a ‘database’ for the event-driven world

Thank you @benstopford Book: https://www.confluent.io/designing-event-driven-systems