~/conference

Observability Summit Europe 2026

date:2026-10-05
location:Prague, Czechia

observability engineers, platform engineers, site reliability engineers, telemetry engineers, cloud native architects, security engineers

OpenTelemetry collector; eBPF profiler; GPU observability for Kubernetes; software carbon intensity; semantic conventions and sidecars; span links and supertraces; OTel-Arrow vs Fluent Bit vs Collector; telemetry testing with Weaver; kernel user_events telemetry; OpenTelemetry metrics performance

titleauthorlength
Beyond Software Estimates: Measured Per-Pod...Jedrzej Wasiukiewicz, Markus Lehtonen27m
A Kubernetes NRI plugin plus a Go resctrl library reads Linux 7.0 resctrl group counters and Intel AET core energy and activity metrics, binds each container…
Talk on
OpenTelemetry Beyond Observability: Security, CI/CD, and Network ManagementAntoine Toulme23m
Claims that OpenTelemetry becomes a backbone for security, CI/CD and networking are supported with concrete work: an OCSF connector inside the collector that…
Talk on
The Noisy Neighbour Inside Your GPU:...Sagar Utekar, Oshi Gupta23m
Continuous batching (vLLM/TGI) fuses strangers on the same GPU step so a 4,000‑token reply can inflate everyone’s inter‑token latency; KV cache for a 70B model…
Talk on
Carbon as a First-Class Signal: Designing a Standard for...Diana Todea18m
OpenTelemetry and the Green Software Foundation map Software Carbon Intensity (SCI, ISO standard since 2024) into telemetry by breaking SCI’s formula (E energy…
Talk on
Where Did the CPU Go? Building the...Nayef Ghattas06m
A Datadog-built OpenTelemetry eBPF profiler runs host-wide, samples all processes with <1% overhead, and finds a 70% CPU increase after an OS upgrade by…
Talk on
Runtime Context: The Missing...Ari Recht07m
Claim that current observability economics and shape break for AI driven operations: vendors charging per host, per gigabyte, per log line collide with a…
Talk on
Welcome + Opening RemarksHenrik Rexed, Iris Dyrmishi07m
First Observability Summit Europe opens with logistical specifics: OpenTelemetry graduated gets an applause, sponsors named are DataDog, Dynatrace, Honeycomb…
Talk on
Vendor-Neutral GPU Observability for KubernetesReza Jelveh20m
GPU observability unifies vendor metrics by instrumenting scheduling rather than scraping drivers: HAMi records scheduling-layer instrumentation and reports…
Talk on
From Schema to Shipping Data, Making...Christos Markou, Pablo Baeyens24m
OpenTelemetry teams describe concrete stability work: migrating Kubelet stats receiver metric names over 10 releases using per-component feature gates…
Talk on
At Engineering Ingegneria Informatica, we...Clemente Biondo12m
An internal developer portal for 5,000 developers wires an LLM agent to live telemetry over MCP and queries an OpenTelemetry stack (Grafana plus Mimir Loki…
Talk on
Joining The Dots: How Span Links And Fake Traces...Alix Klingenberg, Cail Young20m
Engineers build supertraces to represent 12‑day deployments as a single logical trace by emitting a synthetic root span (created via a risky reflection hack in…
Talk on
Telemetry Colosseum: OTel Collector vs. Fluent Bit v5 vs. OTel-Arrow....Henrik Rexed26m
Three agents (OTel Collector, Fluent Bit v5, OTel‑Arrow) run under identical Kubernetes conditions and five trials compare design, signal support…
Talk on
Automating the Toil, Not the Community: AI-Assisted OpenTelemetry...Imma Valls11m
Maintains a Spanish localization and prototypes an AI translation skill that confirms translation state, creates a branch, copies source files, adds link…
Talk on
Do Your Metrics Reflect Customer Experience? Deriving Edge Metrics...David Hayden25m
Turns raw Cloudflare and Envoy Gateway logs into Prometheus-style RED metrics in real time using OpenTelemetry Collector and the signal-to-metrics connector…
Talk on
When Mobile Networks Become Microservices: Tracing the 5G...Robbe De Helt08m
Open5GS is instrumented three ways: manual SDK instrumentation, Istio Ambient service mesh, and eBPF (Grafana Beyla); manual and mesh give equivalent coverage…
Talk on
From Chaos to Consistency: OTel Sidecars and Semantic Conventions in...Karin Fürst11m
Company deploys one OpenTelemetry Collector sidecar per component across ~2,000 Azure subscriptions and 1,700 VM scale sets to get a single pane of glass and…
Talk on
Release, Observe, Decide: OpenFeature...Sean O'Dell05m
A concrete example traces one feature flag from rollout to observed outcome: use OpenFeature evaluations to flag a small cohort, attach OpenTelemetry…
Talk on
Trace Context Beyond HTTP:...Robert Pająk, Alan Clucas23m
Claim: OpenTelemetry environment-variable carriers let trace context cross process boundaries so CI jobs, CLI tools, containers and subprocesses join a single…
Talk on
Testing Your Telemetry Output With Opentelemetry WeaverRiccardo Magliocchetti11m
OpenTelemetry Weaver validates exported telemetry by running an embedded OTLP gRPC server and checking against a registry; a recorded demo points an…
Talk on
What OpenTelemetry Users Taught Us About...Juliano Costa, Johanna Öjeling22m
OpenTelemetry users report that getting to production — not hello world — is the hard part, so the SIG collects concrete configs and patterns: Mastodon runs a…
Talk on
Application Telemetry Through the Kernel With Linux user_eventsLalit Kumar Bhasin23m
Linux user_events lets instrumented apps register tracepoints and write structured logs into per-CPU kernel/perf buffers so a listener can drain CPU-local…
Talk on
Mentorship in the Age of AI: A Repeatable...Severin Neumann, Alyssa Wright26m
A 10‑week cohort runs 48 engineers with seven OpenTelemetry maintainers and produces 55 PRs merged across 11 repos, a participant who becomes the 2nd largest…
Talk on
Just Tune It: Closed Loop Telemetry With MCP, AI Agents, and OpAMPPanos Tsilopoulos26m
Nike runs 100,000 OpenTelemetry Collectors behind an enterprise OpAMP control plane and wires MCP so AI agents can read telemetry and fleet state, draft MCP…
Talk on
Detecting the Dip: Building Observability Around...Karina Celis Jones23m
Built an anomaly-alert system on Prometheus metrics that predicts per-client traffic using a median of three days (due to three‑day retention) plus a trend…
Talk on
Closing RemarksHenrik Rexed, Iris Dyrmishi01m
This announces the end of Observability Summit Europe 2026 and confirms it as the first edition in Prague, thanking sponsors, volunteers and CNCF staff who…
Talk
It Worked Because You Showed Up: Thank...Alyssa Wright02m
A 10-week mentorship pairs 48 Bloomberg technologists with OpenTelemetry maintainers, logging 842 hours and producing 118 pull requests, of which 70 merged…
Talk on
Your Agent Did What? Forensic...Adriana Villela, Kasper Borg Nissen25m
Claim: OpenTelemetry becomes the single convergent standard and the genainormalizer processor at the collector layer bridges incompatible frameworks. Demo…
Talk on
From Paths to Dimensions: How We...Ioannis Androulidakis, Maris Elsins27m
Booking.com migrates ~20 million Graphite time series from a heterogeneous stack to Grafana Mimir using double writing and a runtime routing layer; it deploys…
Talk on
Poisoning the Well: Security Lessons from Vulnerability Research...Shuva Jyoti Kar, Aditi Gupta17m
Vulnerability research finds critical flaws in real telemetry stacks and demonstrates practical fixes: OpenTelemetry Collector and Kafka deployments allowed…
Talk on
OpenTelemetry Metrics Just Got 25× FasterCijo Thomas10m
OpenTelemetry drops Counter recording cost from about 50 ns (with three attributes) to under 2 nanoseconds by pre-resolving attribute sets into bound…
Talk on
~/search
type to search 6,518 conferences, 143,918 speakers and 192,133 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
← →
on a talk: 5 seconds back / forward
0–9
on a talk: jump to 0%–90% of the recording
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.