.png)
3 months ago
1
What's New
Run 5GB/s of Kafka workloads for $50 per hour
Ursa, the data streaming engine powering StreamNative’s managed Kafka service, can sustain 5GB/s Kafka workloads at just 5% of the cost of traditional streaming engines like Redpanda.
Native integration with Snowflake Open Catalog
Snowflake's Open Catalog, a fully managed service for Apache Polaris™, is an open-source catalog enabling secure, centralized access to Iceberg tables across REST-compatible query engines. It is now natively integrated with StreamNative Ursa Engine.
[Live Streaming] Ursa GA announcement
Ursa is now generally available on AWS! You can now run production-ready workloads on the Ursa Engine and take advantage of its leaderless architecture, offering significant cost savings compared to other Kafka alternatives.
Unlock Next-Level Data Streaming with Ursa.
Scale Faster. Run Smarter. Deploy Cheaper. Seamlessly integrate Apache Kafka and Apache Pulsar.
What Ursa means for your business
Kafka API-Compatible
No application rewrite
100% compatible with Kafka API. All your existing Kafka clients and tools work out-of-the-box, now backed by a more cost-efficient and scalable data streaming engine.
No Keeper
Less management
No more time sunk into managing ZooKeeper clusters. Ursa’s leaderless architecture eliminates the need for ZooKeeper, simplifying operations in containerized environments like Kubernetes.
Lakehouse First
Lower costs
Ursa writes data directly to low-cost object storage in open table formats – so all your streaming data is immediately available as Iceberg/Delta Lake tables for querying. No more managing tiers of brokers or extra storage nodes.
.svg)
Empower Your Entire Business with Data Streaming and AI
Kafka Protocol
The de facto streaming protocol, reborn with 10x performance powered by the Ursa Engine
Pulsar Protocol
The unified protocol for message queuing and data streaming. One copy of data with flexible subscriptions—supporting Exclusive, Failover, Shared, and Key_Shared modes.
.svg)
Multi-tenancy allows multiple teams, departments, or customers to share the same infrastructure while keeping data, resources, and workloads securely isolated. With fine-grained access control and resource quotas, it ensures optimal performance, scalability, and cost efficiency—empowering organizations to streamline operations without compromising security or stability.
Ensure low-latency streaming to optimize GenAI inference pipelines.
Integration of streaming and batch data over a single storage layer
Elimination of dependency on specific compute nodes (leaderless), allowing greater flexibility
Support for various data access methods, enabling both real-time and batch processing from the same data set
Ursa Phases
Lakehouse storage
Public preview
2
No ZooKeeper
Public preview
3
Ursa Capabilities
Leaderless Architecture
Decouples metadata (index) from data, enabling true multi-writer & multi-reader concurrency for seamless scalability.
Lakehouse-Native Storage
Stores data directly in Lakehouse tables, eliminating redundant data transfers between Kafka and your Lakehouse.
Integrated Data Catalogs
Provides instant data discovery & unified governance by seamlessly integrating with data catalogs.
Multi-Protocol Support
Fully Kafka API-compatible, allowing direct real-time streaming from Kafka and Pulsar into Lakehouses without modifications.
Discover the benefits of URSA, a Kafka API-compatible data streaming engine built on the lakehouse. Get insights on optimizing your data streaming with URSA!
