G2

Bing L.

Sr. Cloud Infrastructure Engineer

Mid-Market (51-1000 emp.)

8/11/2026

"Redpanda Simplifies Streaming: Single Binary, Great Console, and Strong Support"

5/5

What do you like best about Redpanda Streaming?

The biggest win has been operational: Redpanda is a single binary with no JVM and no ZooKeeper/KRaft quorum to manage separately. Our previous Kafka cluster needed brokers plus ZooKeeper nodes.

Integrations: Kafka API compatibility was the deciding factor.

UI/UX: Redpanda Console gets used daily to check consumer lag, inspect messages on a topic, and confirm schema registry entries without dropping to rpk. Being able to hand a non-engineer a URL to see whether events are flowing has cut a real amount of Slack back-and-forth.

Performance: p99 produce latency sits sustained. Predictability matters more to us than peak throughput.

Pricing / ROI: Fewer, smaller nodes for the same workload dropped our streaming infra spend.

Support / Onboarding: Always respond fast, shared a Slack channel with high quality of the migration help.

AI/Intelligence: Haven't used too much but will definitely try something out. Review collected by and hosted on G2.com.

What do you dislike about Redpanda Streaming?

Observability: Metrics are Prometheus-native and thorough, but mapping them onto the Kafka dashboards and alerts we already had took real effort , names and semantics don't line up one-to-one. Review collected by and hosted on G2.com.

What problems is Redpanda Streaming solving and how is that benefiting you?

We use Redpanda as the ingest and fan-out backbone for high-volume telemetry from fleet devices.

The problem it solved was operational cost per unit of throughput. On self-managed Kafka we were carrying 12 brokers plus a coordination quorum, and a meaningful share of my team's time went to broker tuning, GC pauses, and capacity headroom we kept purely to absorb spikes. That was engineering time spent on the transport layer rather than on the product sitting on top of it.

Since moving to Redpanda we run 12 nodes for the same volume, latency is predictable enough, and cluster maintenance has gone from a recurring task to something I mostly don't think about. Review collected by and hosted on G2.com.