Kafka Summit London 2022: The Full Recap

It’s official: Kafka Summit is back! Technically, it never went away—it just went online. But this week in London, Kafka Summit returned in all its glory to welcome over 1,200 people—real human beings—back to this in-person conference. And I can tell you: it was glorious! Conversations were had, talks were enjoyed, and questions were answered, all as the community convened for two days of talks about Apache Kafka^®.

Ben Stopford MC’d the keynote sessions with aplomb

Day 1 keynote

Original Kafka co-creator and Confluent CEO Jay Kreps took to the stage to discuss where Kafka and event streaming is at, and where it’s heading next. He laid out the five principles under the umbrella of modern data flow:

Some of the principles will have been familiar to many already—such as the fact that streaming is central to so many successful systems because life doesn’t happen in batch. Others brought in concepts from other areas of software development, including the important ideas of decentralization (such as seen with the concept of data mesh), and the importance of pipelines being developer orientated and declarative.

After this, we saw a cool demo of these modern data flow concepts in action, built around the idea of an airline and its decentralized data model and showing some of the tooling that’s now available. You can try it out for yourself using the airline demo recipe.

Next up on stage was Avi Perez, head of backend engineering at Wix.com. He talked about how he and his team are using event streaming to power a full 7% of the internet’s websites through Wix’s no-code web development platform. Their initial Kafka use turned into Confluent Cloud adoption, allowing Wix teams to maintain 20 clusters across two public clouds and three continents connecting 2400 microservices with 70 billion business events every day.

Jay Kreps rounded off the keynote by announcing Current 2022: The Next Generation of Kafka Summit. The inaugural Current will be held in Austin October 4-5, 2022, and you can read more about it in Ben Stopford’s blog. The Call for Papers is already open.

You can watch the full keynote below.

Breakout sessions

As I sat in the sessions, I realized that there is something unique about hearing a talk in person. It’s not just the fact it’s in real life and not a tab on your browser to zone out of. It’s the people near you, the feeling of interest and learning all around you. The muttered exclamations at a cool demo, the whispered nods to colleagues at a great idea to take back and implement in their own system.

Attendees at Kafka Summit were kept busy with 60 sessions over four parallel tracks, ranging from Disaster Recovery Options Running Apache Kafka in Kubernetes to Let’s Make Your CFO Happy: A Practical Guide for Kafka Cost Reduction. Many of the sessions were recorded and will be available online in the coming weeks.

You can check out the #kafkasummit tag on Twitter for lots of notes and commentary on sessions that people watched. Here are just a handful of highlights of the sessions that I attended.

• Antón Rodríguez did a great talk with a flawless live demo on a tool called Pixie which uses eBPF to do observability on Apache Kafka. Anton even modeled his Confluent Community Catalyst cape for us afterwards 😄
  

  #kafkasummit @antonmry using @pixie_run to show @apachekafka consumer lag pic.twitter.com/JwF0h6Qsal

  — Robin Moffatt 🍻🏃🥓 (@rmoff) April 26, 2022

   

• Andrea Spina talked about streaming ML—I didn’t catch all of his talk but definitely need to watch the recording once it’s out.
  

  Fascinating talk from Andrea Spina about streaming ML at #kafkasummit pic.twitter.com/QhLUuRsjo5

  — Robin Moffatt 🍻🏃🥓 (@rmoff) April 25, 2022

   

• The talk from Simona Meriam was fascinating in that it dug into one of the tricky problems when the worlds of batch and streaming collide—when is it the end of day?
  

  #kafkasummit Simona Meriam talking about handling end of day with streams pic.twitter.com/7hYevTxZS9

  — Robin Moffatt 🍻🏃🥓 (@rmoff) April 25, 2022

   

• My colleague Kris Jenkins did a fun talk about building a text adventure game in Kafka—don’t ask why; just watch the video!
  

  #kafkasummit love this demo from @krisajenkins. pic.twitter.com/0CoAkpiERk

  — Robin Moffatt 🍻🏃🥓 (@rmoff) April 25, 2022

   

Apache Kafka committers panel session

Kris Jenkins closed the summit with a panel of the Apache Kafka committers who were at the conference, discussing future ideas and direction for Kafka.

#kafkasummit @apachekafka committers @cmccabe314 @bbejeck @MickaelMaison @davidjacot @BleeGoldman and John Roesler talk about what they see coming for @apachekafka pic.twitter.com/rmKqw106AN

— Robin Moffatt 🍻🏃🥓 (@rmoff) April 26, 2022

 

As well as the committers themselves, there was a nice touch at the Kafka Summit party of a feature wall of some of the many contributors to Kafka:

#kafkasummit party with a cool @apachekafka contributors screen pic.twitter.com/YDsuXvfOkp

— Robin Moffatt 🍻🏃🥓 (@rmoff) April 25, 2022

 

Remembering what in-person conferences are all about

The sessions were great, the keynotes inspirational, yada yada…but what was best about Kafka Summit London 2022? THE PEOPLE. The real people. It was truly wonderful to chat to people, make new friends, reacquaint with old ones, all around the common theme of Kafka but generally just an interest in technology, event streaming—and sharing ideas.

Some #streamingselfie pics from last night, great to catch up with everyone 😁 pic.twitter.com/K5RyOwLmVP

— Kate Stanley (@KateStanley91) April 26, 2022

 

See you in Austin

Current 2022: The Next Generation of Kafka Summit is only five months away! I hope to see you there—and I hope the great talks you heard at Kafka Summit London will have inspired you to consider submitting one of your own.