Confluent Cloud is the industry's only cloud-native service for data streaming at scale built around Apache Kafka. Completely designed for the cloud instead of simply installing Kafka on cloud infrastructure, Confluent provides a complete, fully managed experience for data streaming, processing, and analytics across public and private clouds.
The cloud-native Kafka journey is one that goes beyond provisioning and spans all the way from sizing to scaling. While a cloud-hosted service can simplify your Kafka provisioning, it will still leave you stranded with manual operations and infrastructure monitoring thereafter.
Your choice of a managed Kafka service will either truly enable your organization to exploit all the benefits of Kafka, or decrease developer productivity and delay time to market due to unplanned manual operations.
| Cloud-native experience with Confluent | Cloud-hosted experience with other services | |
|---|---|---|
| Sizing | Throughput-based Eliminate cumbersome performance testing and capacity planning with cluster sizing based on your streaming requirements such as throughput. At the same time, you can reduce infrastructure costs with unlimited cluster-level storage, elastically scalable, scale-to-zero pricing clusters where you only pay for usage versus provisioned infrastructure. |
Broker-based Allocate time and technical resources to run multiple performance tests to pick broker or instance types and count. Pricing is based on infrastructure components provisioned for both compute and storage, for every cluster, even for those that may have low throughput during development. |
| Provision | Self-serve, on-demand Provision Kafka clusters along with any other Confluent components including Schema Registry, connectors, & stream processing with Flink. |
Self-serve, on-demand Provision Kafka clusters only. |
| Infra Monitoring | Confluent proactive monitoring Stay focused on app development with proactive cluster monitoring and maintenance from the Kafka experts. Additionally, with Infinite Storage you can enable unlimited cluster-level use cases while simplifying capacity planning and reducing risk of disk space-related failures. |
Manual monitoring Assign resources to monitor broker metrics such as CPU utilization to proactively manage cluster performance while continuously monitoring and managing alerts for disk space to prevent failures due to storage capacity. |
| Topic monitoring | Pre-aggregated and free metrics Gain the most valuable insights about your applications at no additional cost with access to key metrics aggregated at the topic and cluster level with Data flow. Aggregated metrics can be consumed by your third party monitoring service of choice using Metrics API. |
Per topic per broker metrics cost extra Pay to consume and manually aggregate per-broker and per-topic-per-broker level metrics to monitor topic and cluster level usage. Users must also maintain metric processing and aggregation logic to accurately show data after scaling events when partitions are rebalanced across different brokers. |
| Upgrades | Always on latest version Zero intervention as part of rolling upgrades to latest stable Kafka version that includes strategic patches ahead of scheduled Apache releases. Upgrades are non-disruptive to maintain SLA availability. |
Limited version support Manually trigger upgrades once major versions are supported after a scheduled Apache release. During the upgrade process, cluster availability is also a user responsibility. |
| Vulnerability Patches | Proactive fixes Stream confidently and reliably with Kafka experts that proactively address known bugs and vulnerabilities and resolve even the most complicated Kafka issues. |
Not Available Failures due to software excluded from uptime SLAs. |
| Cluster expansions | Elastic scalability Automatic resource allocation to your cluster to manage consumer lag as throughput scales up or down with self-balancing clusters. With Infinite Storage, you can eliminate over-provisioning of cluster compute while increasing topic retention. |
Add brokers without data balancing Manual data rebalancing process required using service provider tools or third party tools such as Cruise Control after brokers are added to any cluster. Storage per broker/instance limitations force users to either overpay for compute or force exporting data out of Kafka for long data retention use cases. |
| Connectors | Pre-built and fully managed Accelerate integration to modern and legacy services across on-premises and public clouds with a continuously growing portfolio of +120 source and sink connectors. With fully managed connectors, you can reduce the operational burden of provisioning, managing and supporting additional infrastructure for integrations. |
Self develop & manage Slow down delivery timelines with non-repeatable integrations to data services. Operational effort is increased even when using community-built connectors as users self-manage additional connect clusters without support. |
| Non-Java clients | Confluent-supported Enable developer velocity and make Kafka widely accessible to applications and services with a wide variety of battle-tested clients for C, Java, .Net, Go, Python and more. |
Self-supported Leverage self-built or community built clients without any technical support from service provider. |
| Support | Committer-driven expertise Expert 24x7 support engineers have solved tens of thousands of Kafka-related issues that become commits to the open-source project. Accelerate time to market even further with expert guidance from the Professional Services team. |
Limited expertise Limited experience supporting and maintaining Kafka and its ecosystem. |
| Environments | Freedom of choice Consistent cloud native experience across AWS, Azure and Google Cloud with option to subscribe directly through their respective marketplaces for simplified billing. Ability to extend event-driven architecture to on-premises or private cloud environments using Confluent Platform, our self-managed software. |
Limited Service limited to a single cloud provider and/or lacks a self-managed software version for simplified Kafka operations on-premises. |
| Ecosystem | Complete Confluent offers way more than just a Kafka service with a complete ecosystem including schema management and stream processing. Developers can maintain application compatibility with fully managed Schema Registry and develop real-time ETL pipelines with fully managed Flink. |
Limited Apache Kafka clusters only. |