Announcing IBM FlashSystem for Confluent Platform: Lab-Tested Enterprise Storage for Real-Time Data and AI, On-Premises

Blog written by Vincent Hsu, Fellow, CTO and VP, IBM Storage, IBM, published October 28, 2026

IBM Storage and Confluent are pleased to announce that IBM FlashSystem has been lab-tested as persistent storage for Confluent Platform on Red Hat OpenShift. Organizations that stream data on-premises can now run Confluent Platform on IBM FlashSystem with confidence, using the standard Kubernetes storage interfaces they already rely on.

Why it matters
Every payment, sensor reading and customer interaction is an event. Confluent, founded by the original creators of Apache Kafka, connects those events across applications, databases and systems of record in real time, adding enterprise security, governance, connectors and Apache Flink stream processing to open-source Kafka.

For AI, that real-time foundation is essential. Models and agents are only as good as the data they act on, and streaming gives them fresh, governed context the moment it is created. At the same time, banks, insurers, healthcare providers, telecom operators and public-sector agencies must often keep that data in their own data centers. Confluent Platform, deployed with Confluent for Kubernetes on Kubernetes and Red Hat OpenShift, gives them enterprise data streaming on infrastructure they control.

Your platform, your storage choice
Confluent Platform is built to run on the infrastructure customers already trust. Each Kafka broker requests its storage through a standard Kubernetes StorageClass, so customers choose the enterprise storage that meets their own performance, resilience and operational standards. With today’s announcement, IBM FlashSystem joins that choice as a tested option.

What IBM FlashSystem brings
IBM FlashSystem is IBM’s all-flash enterprise block storage family, trusted for core banking, databases such as Oracle, Db2 and SAP HANA, virtualized estates and containerized applications on OpenShift. For Confluent Platform deployments, it delivers:

1. Consistent performance: high throughput and low latency from all-flash FlashCore Modules
2. Kubernetes-native operations: the IBM block storage CSI driver provisions a volume per broker on demand and expands it online
3. Performance isolation: volume throttling keeps neighbouring workloads from affecting brokers
4. Security: AES-256 encryption of data at rest, alongside TLS and Confluent client-side field-level encryption
5. Cyber resilience: AI-based ransomware detection and immutable Safeguarded Copy snapshots for a clean recovery point
6. Availability: Policy-Based High Availability across two sites and Policy-Based Replication for disaster recovery, complementing Confluent Cluster Linking
7. Mobility: FlashSystem Grid moves volumes between arrays without disruption

FlashSystem connects over both Fibre Channel and Ethernet (iSCSI), so customers can use the storage network they already run.

Tested in the lab
IBM tested Confluent Platform, deployed with Confluent for Kubernetes on Red Hat OpenShift, with FlashSystem as the broker storage. Testing covered dynamic provisioning, online volume expansion, ransomware detection and non-disruptive migration, with no changes to how Confluent Platform is deployed.

Get started today
Customers can deploy Confluent Platform on IBM FlashSystem today, using Confluent for Kubernetes and the IBM block storage CSI driver.