Actian recently announced Apache Spark support in an updated version of its Vector in Hadoop (VectorH) tool. The solution takes advantage of vectorized processing and in-memory acceleration for Hadoop data stores. The provider’s family of Vector analytic database offerings work on single server, cluster and hybrid transactional/analytic environments both on-prem and in the cloud.
Analytic functionality included in VectorH is aimed at developers and data scientists that utilize Apache Spark. Some of Spark’s most noteworthy uses include machine learning, streaming, and prediction. VectorH in conjunction with Spark acts as a production platform that provides security, data updates in real-time, resource management, and query optimizers.
A few highlights of Actian’s latest and greatest include:
- Data updates in real-time: According to the vendor, the tool can process data updates without downgrade on performance
- Native query access: Native Spark support provides optimized access to Hadoop file formats including Parquet and ORC. Direct access will include the ability to perform SQL joins across tables.
- Native DataFrame support: Provides a direct connection to Spark features via DataFrames, which enables faster query execution in SparkSQL and Spark R apps.
In a press statement, Actian’s CEO Rohit De Souza explained: “Customers facing the challenges of today’s dynamic, hybrid data environments are demanding faster, more nimble solutions to activate that data. Actian now offers the power of VectorH through Spark to support a diverse range of file formats and workloads including machine learning, delivering performance and operational advantages over open source and proprietary alternatives.”
The new version will be released in late October.
Latest posts by Timothy King (see all)
- The 4 Best Electronic Data Interchange (EDI) Software Tools to Consider - July 16, 2018
- The 28 Best Data Integration Software Tools for 2018 - July 13, 2018
- The 9 Best On-Premise Data Integration Software Tools to Consider - July 11, 2018