On Thursday, April 14, 2022, the C2C DACH region hosted a powerful and informative event with guest Joel Goodman, a staff cloud architect at DoiT International, focused on Cloud Workflows, a serverless offering from Google Cloud. Thomas Hug ( @tom ), one of our DACH Team Leaders, served as moderator of the session. Read on below to review the key takeaways from this in-depth introduction to this dynamic product.
30 Minutes in 30 seconds
- (3:47) Joel Goodman began his presentation by explaining why a Google Cloud user should consider Workflows. The more services a user has talking to each other at once, the harder they are to manage, and the more tedious the process of sending events to a pipeline becomes.
- (5:01) Goodman compared Workflow to an orchestrator: a central process that executes the workflow from start to finish.
- (5:26) Next, Goodman gave an overview of Workflows and its capabilities, and what writing a workflow looks like.
- (7:00) To provide some examples of use cases for Workflows, Goodman brought up microservice orchestration, continuous integration and deployment (although he admitted he wouldn't use it for heavier things), transactional consistency, ETL and Data pipelines (although he acknowledges that Workflows would be a better way to start for light data pipelines, and that for more complex needs there are a number of other tools available that would be more suitable), and long-running workflows.
- (9:00) To give attendees a demo of Workflows, Goodman used the example of a mechanic who runs an application whose users send pictures of their vehicle. The microservice orchestration in this case is as follows: save the image to Google Cloud Storage, extract the license plate number from the image, look up the vehicle's information, save the information to BigQuery, and finally email the vehicle image to the mechanic with the information he needs.
- (10:44) Next Goodman listed some workflow design requirements: It has to be cheap, it has to focus on business logic and not infrastructure, it has to scale up and down with customer demand, and it has to be reliable and allow for easy troubleshooting.
- (11:07) Goodman also expanded on his design decisions - - Cloud Run for microservices and Cloud Workflows for orchestration - - and analyzed the managed and external services and the microservices required.
- (11:58) To break everything down further, Goodman explained the specific uses for the web app, the license plate reader, and the notification service.
- (12:25) Goodman next provided an extensive analysis of the differences between orchestration and choreography.
- (16:40) Finally, Goodman gave attendees a demonstration of the application's front end, the submission of the image, and the process in the background.
- (20:56) For the rest of the session, Goodman fielded questions from C2C members hoping to implement Cloud Workflows for their own services.
Extra Credit:
Looking to get more involved with our DACH community? Come to our in-person event in Munich on May 18, 2022. This session will cover how MediaMarktSaturn built its Data Mesh, and why this solution is such a game-changer. Attendees will also hear from our partner AMD about how they are making their industry leading AMD EPYC
processors available on Google Cloud, and how UberCloud is helping organizations run their simulation tools using HPC application containers.
Join us and these amazing speakers as they share their journeys and business outcomes, and how they have overcome their technical and business challenges:
-
Fabian Seitz, Group Product Manager, MediaMarkt Saturn
-
Pawel Walczysko, Cloud Solution Specialist
-
Wolfgang Gentzsch, President, UberCloud
-
Daniel Gruber, Director of Architecture, UberCloud
Sign up below today!
