Data and event streaming

# Apache Kafka for cloud data streaming

Deploy Kafka in an isolated container with private networking. Size brokers and persistent storage for your workload within the limits of your zone package.

[Get started](https://panel.deplomo.com/signup)[View in marketplace](/marketplace)

## Apache Kafka features on Deplomo Cloud

Use Apache Kafka for event streaming with a container-based cloud deployment.

### Very high throughput

Kafka can process high message volumes for data streaming workloads; throughput and latency depend on your configuration and workload.

### High fault tolerance and data redundancy

Replication across independent partitions can help Kafka recover from some hardware issues. Data durability depends on replication, retention, and cluster configuration.

### Log retention and replay

Kafka stores records on persistent storage according to the configured retention period, allowing consumers to replay events while they remain available.

### Processing event streams

Kafka’s event-driven architecture can support analytics pipelines, monitoring workflows, and asynchronous communication between microservices.

### Kafka deployment and configuration

Deploy a Kafka broker without setting up Linux repositories manually. Review the network and broker settings for your environment.

### Scale resources as your project grows

Plan Kafka CPU and RAM allocations within your zone package limits as the incoming data volume grows.

## Frequently asked questions about Apache Kafka

Answers to common questions from system architects and data engineers about Kafka, its uses, and deployment:

What is Kafka, and what problems does it address in software architecture?

How does Apache Kafka differ from message brokers such as RabbitMQ?

How do you connect applications to Kafka in the cloud?

Is ZooKeeper required to run Apache Kafka on Deplomo?

How is Kafka service pricing calculated on Deplomo Cloud?

## A comprehensive guide to Apache Kafka; Streaming architecture, big data and real-time processing

On Deplomo Cloud, Apache Kafka runs as a single-node KRaft service for private-network use through the  [Cloud containers](/products/containers)  deployment.

### Apache Kafka architecture: topics, partitions, and consumer groups

Kafka organizes records in topics and partitions. Producers write events to partitions, and consumers read them using offsets. Consumer groups can distribute processing across nodes, subject to cluster capacity.

<table class="w-full min-w-[600px] border-collapse text-sm"><thead><tr class="border-b border-border/80 bg-background/50 text-foreground"><th class="p-3 text-start font-bold">Comparison</th><th class="p-3 text-start font-bold text-primary">Apache Kafka on Deplomo</th><th class="p-3 text-start font-bold">Manual setup on a VPS</th><th class="p-3 text-start font-bold">Traditional classic message queues</th></tr></thead><tbody class="divide-y divide-border/60"><tr><td class="p-3 font-semibold text-foreground">Setup approach</td><td class="p-3 text-primary font-medium">Container-based deployment; setup depends on template and configuration</td><td class="p-3">Setup steps vary by software and environment</td><td class="p-3">Depends on the software and tools used</td></tr><tr><td class="p-3 font-semibold text-foreground">Throughput</td><td class="p-3 text-primary font-medium">Throughput depends on broker resources and partition settings</td><td class="p-3">Depends on resources and server partition settings</td><td class="p-3">Depends on broker configuration and workload</td></tr><tr><td class="p-3 font-semibold text-foreground">Persistence and playback of data</td><td class="p-3 text-primary font-medium">Log-based storage with configurable retention</td><td class="p-3">Persistence and cleanup depend on configuration</td><td class="p-3">Delivery and message retention depend on queue configuration</td></tr><tr><td class="p-3 font-semibold text-foreground">Networking and access</td><td class="p-3 text-primary font-medium">Configured for private-network use; check access settings before connecting</td><td class="p-3">Requirements depend on the deployment and firewall configuration</td><td class="p-3">Access isolation depends on network configuration</td></tr><tr><td class="p-3 font-semibold text-foreground">Cost model</td><td class="p-3 text-primary font-medium">Fixed monthly EUR package pricing</td><td class="p-3">Pricing depends on the provider and selected capacity</td><td class="p-3">Pricing depends on the hosting or software option</td></tr></tbody></table>

### Kafka and message brokers: when to consider RabbitMQ or NATS

For task queues and flexible routing patterns, consider [rabbitmq](/marketplace/apps/rabbitmq)  can be an option. For lightweight messaging between services, consider [nats server](/marketplace/apps/nats)  depending on your requirements. Kafka is designed for event streaming and retained logs; compare each system’s features against your workload.

### Build microservices with modern frameworks

Kafka client libraries are available for languages such as [Go hosting](/products/go) . In Go, Sarama and confluent-kafka-go are examples; Python, Java (including Spring Boot), and Node.js also have Kafka clients. Latency depends on the application and deployment.

### Kafka deployment and topic management

To plan Kafka capacity, consider the number of partitions and active consumers for each topic. Choose message keys carefully when related events need to stay ordered within a partition, and set a retention period that fits your storage capacity and replay needs.

### Evaluate Kafka with the package that fits your project

Kafka uses the CPU, memory, and storage included in your selected monthly package. International service creation is currently paused; no promotional credit is included.

## Deploy Apache Kafka

Plan your data-processing service without manually configuring a Linux server. Choose a fixed monthly zone package that meets its resource requirements.

[Apache Kafka deployment](https://panel.deplomo.com/signup)