RabbitMQ on Kubernetes Performance and Operations Training

RabbitMQ on Kubernetes Performance and Operations

A focused, advanced course for engineers who run RabbitMQ on Kubernetes. The goal is to improve cluster setup, raise throughput, reduce latency, and harden reliability using proven patterns for scheduling, storage, networking, and broker tuning.

What will you learn?

You will validate your deployment pattern, baseline performance, and implement targeted optimizations in RabbitMQ and Kubernetes. You will apply observability, scaling, and reliability techniques to meet concrete SLOs while keeping operations safe during upgrades and failures.

After this training you will be confident in:

Choosing and hardening a deployment approach on Kubernetes and aligning it to failure domains
Tuning queues, connections, channels, confirms, and prefetch for predictable throughput and latency
Optimizing storage, networking, and resource requests or limits for stable performance
Operating quorum and stream queues, handling failures, and planning controlled rollouts
Using metrics and tracing to detect bottlenecks and prevent regressions
Securing traffic and access with TLS, least privilege, and network policies

Requirements:

Strong familiarity with RabbitMQ fundamentals and Kubernetes basics
Comfort with kubectl, Helm or the RabbitMQ Cluster Operator, and container registries
Access to a non-production cluster with permissions to create namespaces, StatefulSets, Services, Ingress, and Secrets

Course Outline*:

*We know each team has their own needs and specifications. That is why we can modify the training outline per need.

Module 1: Deployment patterns and cluster architecture

Helm charts vs RabbitMQ Cluster Operator and when to use each
StatefulSets, headless Services, and stable network identities
Node pools, topology spread, anti-affinity, and PodDisruptionBudgets
Version strategy and plugin selection for performance features

Module 2: Storage and durability for predictable throughput

StorageClasses, PVCs, throughput vs IOPS tradeoffs, and filesystem notes
Durable, quorum, and stream queues and their operational implications
Lazy queues for cold traffic, message TTL, and dead-letter routing patterns
Disk alarms, watermark tuning, and safe compaction expectations

Module 3: Networking and connectivity at scale

ClusterIP, headless, and external access via Ingress or LoadBalancer
Connection management: connections vs channels, heartbeats, and TCP keepalives
MTU, kube-proxy modes, and avoiding unnecessary hops
TLS termination choices and impact on latency

Module 4: Baseline performance and broker tuning

Perf-test methodology, representative payloads, and SLO-driven scenarios
Publisher confirms strategies, in-flight limits, and batching
Consumer flow control with basic.qos prefetch and fair dispatch
Memory alarms, garbage collection considerations, and VM resource tuning

Module 5: Reliability engineering on Kubernetes

Quorum queue behavior, leader election, and placement across failure zones
Stream queues for high-throughput append workloads and retention configuration
Drains, reschedules, and graceful shutdown to prevent message loss
Rolling updates and node maintenance with minimal disruption

Module 6: Observability, SLOs, and capacity planning

Prometheus metrics to watch: channels, confirms, acks, unacked, consumers, memory, disk
Grafana dashboards for capacity and saturation signals
Tracing message paths and identifying hot exchanges or bindings
Forecasting capacity and setting alert thresholds that reduce noise

Module 7: Security and governance

TLS between clients and nodes, certificates with cert-manager, and rotation
Least-privilege users, vhosts, and permissions aligned to applications
Kubernetes NetworkPolicies to isolate brokers and client namespaces
Image provenance, supply chain considerations, and secret handling

Module 8: Troubleshooting and continuous improvement

rabbitmq-diagnostics and rabbitmqctl techniques for live systems
Detecting and fixing blocked connections, flow control, and backlogs
Shovel and Federation throughput tuning and backlog recovery
Playbook for incident response, postmortems, and controlled performance experiments

Hands-on learning with expert instructors at your location for organizations.

Level:

Intermediate

Duration:

Hours (days:

)

Training customized to your needs

Immersive hands-on experience in a dedicated setting

*Price can range depending on number of participants, change of outline, location etc.

Get Quote

Master new skills guided by experienced instructors from anywhere.

Level:

Intermediate

Duration:

Hours (days:

)

Training customized to your needs

Reduced training costs

*Price can range depending on number of participants, change of outline, location etc.

Get Quote

You can participate in a Public Course with people from other organisations.

/per trainee

Thanks for the numbers, they could be going to your emails. But they're going to mine... Thanks ;D

Oops! Something went wrong while submitting the form.

Level:

Intermediate

Duration:

Hours (days:

)

Fits ideally for individuals and small groups

Networking opportunities with fellow participants.

*Price can range depending on number of participants, change of outline, location etc.

Get Quote

RabbitMQ on Kubernetes Performance and Operations

What will you learn?

Requirements:

Course Outline*:

Module 1: Deployment patterns and cluster architecture

Module 2: Storage and durability for predictable throughput

Module 3: Networking and connectivity at scale

Module 4: Baseline performance and broker tuning

Module 5: Reliability engineering on Kubernetes

Module 6: Observability, SLOs, and capacity planning

Module 7: Security and governance

Module 8: Troubleshooting and continuous improvement

Get Quote for:

RabbitMQ on Kubernetes Performance and Operations

Start your training experience here!