2 Scaling

dirty south balkan brass band, saxophone alt-country

Listen on 93

Lyrics

[Verse 1]
When your cluster hits the ceiling, time to multiply
Add more brokers to the fleet, watch throughput amplify
Edit your Kafka custom resource, bump the replica count
Strimzi orchestrates the growth, every instance paramount

[Chorus]
Scale up, scale down, brokers join the dance
Partition shuffle follows every scaling chance
Cruise Control and KafkaRebalance hold the key
Connect workers flex and bend with workload's spree

[Verse 2]
Partitions cling to old nodes when you shrink the pool
Reassignment algorithms become your sharpest tool
Leaders migrate, replicas shift to active hosts
Balance the load distribution across remaining posts

[Chorus]
Scale up, scale down, brokers join the dance
Partition shuffle follows every scaling chance
Cruise Control and KafkaRebalance hold the key
Connect workers flex and bend with workload's spree

[Verse 3]
Deploy KafkaRebalance custom resource with care
Cruise Control analyzes metrics hanging in the air
CPU and network bandwidth, disk utilization stats
Generate optimization proposals, iron out the stats

[Bridge]
Connect clusters scale independently from broker nodes
Increase worker replicas when connector workloads explode
Task distribution spreads across the worker pool
Parallel processing keeps the data pipeline cool

[Chorus]
Scale up, scale down, brokers join the dance
Partition shuffle follows every scaling chance
Cruise Control and KafkaRebalance hold the key
Connect workers flex and bend with workload's spree

[Outro]
Elastic infrastructure bends but never breaks
Strimzi manages the complexity that scaling makes
From single broker testing to production's mighty roar
Kubernetes orchestration opens every scaling door

← 1 Rolling Updates & Upgrades | 3 Cruise Control for Rebalancing →