The Starburst Enterprise Presto LTS 345-e release includes many significant features that help Starburst customers with new and enhanced connectivity, improved performance, and more robust security. It is primarily geared towards customers leveraging RDBMS databases, SAP HANA and Greenplum, and enhanced security features.
Our foundation is being a single point of access to data anywhere. When customers ask us, “Can you connect to xyz data source?”, the answer of course, is yes. However, it’s more than that. We’re continually developing enhancements and extensions to our connectors that allow faster and more secure access, and deeper integrations. “Connecting” is only the start. With that in mind, this major release combines features that have been contributed back to the open source project as well as curated for Starburst Enterprise Presto customers. To experience this latest release first hand, you can download it here.
{{cta(‘eddbc327-0144-4f85-b4b2-4ce80ebaeed1′,’justifycenter’)}}
The new release includes:
New SAP Hana and Greenplum Connectors:
For customers using SAP HANA and Greenplum, we’re pleased to announce new Starburst Enterprise connectors. The SAP HANA was our most commonly requested connector. The in-memory, column orientated connector allows for fast access and analytics for organizations across many industries looking for insight into their enterprise resource planning data. The SAP HANA connector allows reads, pushdown and dynamic filtering (see below), and password credential passthrough for additional security.
In addition, this release features a new Greenplum connector. Commonly used in the fintech industry, the Greenplum connector includes basic read capabilities, and enhanced performance of parallelism, table statistics, and pushdown and dynamic filtering. Table statistics enables the Greenplum connector to be used with the Presto Cost-Based Optimizer. For security enhancements, the Greenplum connector leverages user impersonation, and Kerberos authentication and credential passthrough.
Enhanced Delta Lake and Kafka Connectors:
Delta Lake CTAS (Create Table as Select)
CTAS is a parallel operation that creates a new table based on the output of a SELECT statement. CTAS is the simplest and fastest way to create and insert data into a table with a single command. Overall, Delta Lake adds reliability to data lakes with ACID transactions. CTAS continues our evolution of enhancements to the Delta Lake connector and is the initial phase of write functionality.
Kafka Write Operations
Apache Kafka continues to gain adoption in the market, and is replacing other Enterprise Service Bus (ESB) tools across industries. With this latest release, users can now write directly to a Kafka topic through Presto.
Performance Improvements for RDBMS Connectors
Our RDBMS connectors get a giant performance boost with LTS 345. On industry standard TPC-DS benchmark, we’ve seen results up to 58x faster, and 8x faster on average! (Note these results were perform on Oracle connector for TPC-DS dataset and queries. Mileage may vary depending on workload, connector, and relational database.)
Dynamic Filtering
Dynamic filtering significantly improves the performance of queries with selective joins by avoiding reading of data that would be filtered by join conditions. For example, customers can join data from separate tables (within the same data source, or across data sources) faster, such as a customer dimension in a secure database, and a fact table in a data lake. The results of dynamic filtering are, improving overall query performance, reducing network traffic between Presto and the data source, and reducing the load on the remote data source itself. It’s a win-win-win for Starburst customers.
Aggregate Pushdown
Presto can push down the processing of queries, or parts of queries, into the connected data source. This means that a specific function, or other operation, is passed through to the underlying database or storage system for processing. The data source completes calculations previously done in Presto.
The benefit is getting answers faster because you aren’t moving data around. Thus, reducing compute capacity, and resulting in lower costs. This is incredibly useful for customers federating from RDBMS sources, as it optimizes performance and utilization on Presto and RDBMS.
All frequently used operations are supported (COUNT, MIN, MAX, SUM, AVG, STDDEV, VARIANCE, and more)
Security Improvements
Starburst now has an even stronger enterprise security offering with new support for Privacera, the cloud data governance and security leader founded by the creators of Apache Ranger. This provides enterprises with centralized, secure access to data that includes end-to-end governance and compliance.
Benefits include:
Federated Analytics with Fine-Grained Access Control: Using a single-pane to secure data across multiple sources and automate governance and access policies, Privacera and Starburst Data streamline access to data distributed across data lakes and many other sources, so it can be analyzed by data teams, while being subject to consistent security and governance. Users can query multiple data sources from a single location and ensure sensitive data is protected.
Automated Governance, Compliance, and Improved Productivity: Privacera migrates existing compliance policies to the cloud for use in analytical services from Starburst Enterprise for Presto. This ensure enterprises have consistent governance and compliance policies implemented across their on-premise and cloud infrastructures to meet stringent regulations like GDPR, LGPD, and CCPA. This functionality alleviates manual processes from administrators, because no new policies need to be written – saving enterprises time, money, and resources.
Ease of Use Leveraging Apache Ranger Infrastructure: For seamless deployment, the Privacera and Starburst Enterprise are integrated via a lightweight Apache Ranger plug-in. This allows users to take advantage of the same, trusted capabilities of Apache Ranger without any manual reconfiguration, software updates, or interruption of operations.
Additional open source features: Improved support for ADLS Gen 2, Kerberized and impersonated Hive caching, and Improved Glue metastore access.
For more information on this major release, please reference our release notes and experience our latest release for yourself.
