Name: Grafana & Friends France : édition de novembre chez Elaia
Start: 2025-11-27T18:00:00+01:00
End: 2025-11-27T20:00:00+01:00
Location: Elaia

We're hosting this event with our friends from [Paris Apache Kafka](https://www.meetup.com/paris-apache-kafka-meetup/), showcasing Grafana OSS and Kafka technologies.

📍**Venue:**
Elaia
21 Rue d'Uzès, 75002 Paris, France

🗓 **Agenda:**

* 6:00pm: Doors Open/Welcome, Drinks
* 6:15pm - 7:00pm: S*top Stitching Dashboards: How Kafka + Iceberg Deliver Seamless Time Travel* (Tom from Streambased)
* 7:00pm - 7:45pm: *Building Database Observability: Tales from the Trenches* (Shirley from Grafana)
* 7:45pm - 8:30pm: Food, Additional Q&A, Networking

**Abstracts** -
*Building Database Observability: Tales from the Trenches*: Behind-the-scenes stories from working on Grafana Lab's Database Observability project: unexpected quirks and what we learned along the way. A little technical, a lot of reflection, and hopefully a good story.

*Stop Stitching Dashboards: How Kafka + Iceberg Deliver Seamless Time Travel*
Grafana’s magic comes from seeing what’s happening right now and instantly comparing it to everything that has happened before.\* Real-time data lets teams spot anomalies the moment they emerge.

\* Long-term data reveals whether those anomalies are new, seasonal, or the same gremlins haunting your system every quarter.But actually building this capability? That’s where everything gets messy.Today’s dashboards are cobbled together from two very different worlds:\* Long-term data living in lakes and warehouses
\* Real-time streams blasting through Kafka or similar systemsThese systems rarely fit together cleanly, which forces dashboard developers to wrestle with:\* Differing processing concepts - What does SQL even mean on a stream?
\* Inconsistent governance - Tables vs. message schemas, different owners, different rules
\* Incomplete history - Not everything is kept forever, and you never know what will vanish next
\* Maintenance drift - As pipelines evolve, your ETL always falls behindBut what if there were no separation at all?Join us for a deep dive into a new, unified approach where real-time and historical data live together in a single, seamless dataset. Imagine dashboards powered by one source of truth that stretches from less than one second ago to five, ten, or even fifteen years into the past, without stitching, syncing, or duct-taping systems together.
Using Apache Kafka, Apache Iceberg, and a modern architectural pattern that eliminates the old “batch vs. stream” divide, we’ll show how to:\* Build Grafana dashboards that just work with consistent semantics end-to-end
\* Keep every message forever without drowning in storage costs
\* Query real-time and historical data with the same language, same governance, same everything
\* Escape the ETL death spiral once and for allIf you’ve ever wished your dashboards were both lightning-fast and infinitely deep, this talk will show you how close that future really is.

Grafana Community

Grafana & Friends France

Grafana

Grafana Labs

Technology

Open Source

Performance Monitoring

Network Management & Monitoring

Application Performance Monitoring

Metrics

Site Reliability Engineering (SRE)

Observability

Grafana & Friends France : édition de novembre chez Elaia

Elaia

Partager

Grafana & Friends France

Grafana & Friends France : édition de novembre chez Elaia

Grafana & Friends France

Détails

Sponsors

Les membres s'intéressent également à