Splice Machine Readies New Edition Of Its Distributed SQL Data Platform

The company will ship Splice Machine 3.0 this quarter, offering new workload management, high-availability and security enhancements that better position the next-generation database for mission-critical deployments.


Next-generation database developer Splice Machine is gearing up to ship the next release of its intelligent SQL data platform with expanded functionality, including replication and high-availability capabilities and enhanced security, designed to boost the software’s agility and scalability for mission-critical applications.

Splice Machine 3.0, expected to be available by the end of the first quarter, will also provide new workload management features, extended SQL coverage, improved data science productivity and Kubernetes support.

“It’s all about operationalization,” said CEO Monte Zweben (pictured) in an interview with CRN. The focus of the new release is providing the capabilities needed for the software to “scale in a mission-critical application.”

Sponsored post

[Related: The Big Data 100 2019]

That includes reliability and high performance, and the ability to smoothly allocate resources as needed and handle system failures. And it includes using advanced analytics and machine learning to develop predictive models and inject them directly into operational applications, Zweben said.

Splice Machine is a scale-out SQL relational database that supports ACID (atomicity, consistency, isolation and durability) transactions, in-memory analytics and in-database machine learning. The software is designed to help businesses and organizations modernize their legacy and custom applications and inject them with new operational artificial intelligence and machine-learning capabilities. Unlike most traditional databases, the Splice Machine software is adept at processing both transactions and analytical tasks.

Splice Machine works with a range of solution provider and systems integrator partners including Accenture, Capgemini, Diyotta and OnX Enterprise Solutions. Zweben said partners aren’t just implementing IT, but “selling and generating business outcomes” through mission-critical applications with AI and machine- learning capabilities.

“This release allows [partners] to deliver highly reliable machine learning and AI capabilities in the fabric of the real-time enterprise workflow,” the CEO said of the prime benefit for Splice Machine partners.

Zweben said the new release meets increased demands for agility, which he defined as “the ability to use computational resources at scale when you need them and provide them for the right tasks.” One way the new release of the Splice Machine platform does this is through the support of Kubernetes and Kubernetes infrastructure for the containerization and deployment of microservices.

In workload management, the 3.0 version uses application server queues to isolate workloads to ensure adequate resources are available when multiple queries are running simultaneously. On the SQL coverage side the new release offers full outer join options, enhanced trigger capabilities, and DB2-specific SQL syntax —the latter making it easier to migrate applications from that IBM database. A new “point-in-time” SQL syntax extension allows users to query a database as it existed at some time in the past.

For replication and high availability, capabilities required by many mission-critical applications, the new edition offers asynchronous active-passive replication to automatically keep multiple database clusters in sync to meet rigorous recovery time requirements. The software’s security gets a boost from new schema access restrictions and customized pattern matching for redacting sensitive information. Data science productivity is improved through support for popular Jupyter notebooks and the new MLManager platform for model workflow management.