SAN FRANCISCO, October 30, 2024 -- StreamNative, the cloud-native data streaming company founded by the original creators of Apache Pulsar, today made a series of announcements at this week's Data Streaming Summit 2024, including the public preview of Ursa Engine, designed to address the need for cost-effective, future-proof streaming data platforms.
The Ursa Engine, built upon Apache Pulsar's cloud-native architecture, is a 100% Kafka-compatible engine that is built on data lakehouses. It brings the power of real-time data streaming to data lakehouses and augments it to create a single, unified platform for greater flexibility, scalability, and cost-efficiency. Ursa shifts from the ZooKeeper and BookKeeper-based architecture toward a headless stream storage architecture, using Oxia as a metadata store, Object Storage as data storage and making BookKeeper optional. By leveraging the Ursa Engine, users will:
- Reduce costs associated with scaling streaming data infrastructure
- Eliminate the need for costly data transfers between streaming and batch systems
- Seamlessly integrate real-time and historical data, allowing for more efficient AI workflows
Utilizing Ursa Engine will reduce data streaming infrastructure costs for customers, with 10x cost-savings, making it ideal for a real-time data streaming infrastructure. By bringing data streaming closer to data lakehouses, it is designed to meet the unique demands for the age of AI—whether it's training large models, running real-time AI applications, or handling mission-critical transactional workloads. Ursa Engine is available for public preview on StreamNative Cloud. It's available starting today on BYOC (Bring Your Own Cloud), and will soon be available for Dedicated and Private Cloud.
Universal Linking (UniLink)
Today StreamNative also introduced Universal Linking (UniLink), a service that simplifies how organizations handle data replication and migration between different streaming platforms, offering a more cost-effective approach compared to traditional methods. By leveraging object storage instead of direct network connections, UniLink enables seamless data sharing between Kafka and Pulsar environments, eliminating the traditional barriers between these platforms. This innovation is particularly valuable for organizations managing hybrid cloud environments or those looking to migrate between different streaming platforms without service disruption.
UniLink marks a significant shift in how streaming data can be replicated across different environments, addressing a long-standing challenge in the industry. While traditional replication methods rely on expensive network transfers, UniLink's innovative approach uses cloud storage as an intermediate layer, significantly reducing costs while maintaining data consistency.
UniLink is available today as a Private Preview feature on StreamNative Cloud for the company's early access partners.
Data Streaming Summit
Today's announcements were made at the sold-out Data Streaming Summit in San Francisco, a community that came together to share best practices and discuss the future of streaming technologies. Sponsors include PingCAP, Ververica, Google Cloud, Nexaminds, Snowflake, and more while speakers at the conference include executives from FICO, Uber, Pinterest, Databricks, Iterable and more. The Data Streaming Summit, previously named the Pulsar Summit, has been a semi-annual event since 2020, gathering more than 2,000 technologists from more than 170 countries around the world.
"The data streaming industry is at a pivotal moment as organizations build their AI infrastructure and seek more cost-effective, scalable solutions. With this week's announcements including Ursa Engine, we're fundamentally reimagining data streaming architecture by unifying streaming and batch data processing in a single platform. This innovation allows organizations to reduce their infrastructure costs by up to 90% while maintaining the performance and reliability they need for mission-critical workloads. We're not just improving existing technologies—we're creating the foundation for the next generation of AI-powered data streaming." -- Sijie Guo, CEO and co-founder of StreamNative
For more information:
