Gopherfest 2017 | Event Sourcing – Architectures and Patterns with Matt Ho

Event sourcing and other applications of append-only logs have experienced a renaissance in recent years. From git, to Kafka, to block chains, the use of append-only logs is on the rise.

In this talk, Matt Ho will cover the benefits of event sourcing and how to get started with it using Go. He'll provide the background of event sourcing along with simple illustration of how event-sourcing might be applied to a domain entity.

He'll go in-depth on how event sourcing can address challenges posed by transactional integrity, dual writes, and scaling – issues which are increasing in importance given the rise of REST-based microservices.

#GoLang #Gopherfest

[muhamotto]

Is it possible to share the source code please :) ?

[oguzhanyldrm]

We have all the events at DynamoDB already what is the reason to store also at S3?

[DawidTomkalski]

Is there any reason lambda calls the firehose and then S3 instead of directly to S3?

[ramih]

Thanks alot for the great talk, was very informative and educative. I wanted to ask why did you base64 encode the json object? are there any benefits for doing that before storing the object in dynamodb?

[BrahmaAcharya]

This is a good talk. However, I am not sure why you keep referring that in the Microservices world, I would have to do this and that. Event sourcing and Microservices can go hand in hand. The example that you have shown is about Order information. Well that's a MS. I am sure you would still want to keep your domain's bounded context independent of each other. Also, your consumers of the events themselves could implement tasks which could be again a MS.

[benjaminjones]

Similar principles to redux, very cool!

[mishasawangwan]

just wanted to note: lambda layers make the dependency thing an almost, non-issue.

[_akshaydeep]

cloud is all good and great but i will never let them have my business data.