Data warehouse services (DWS) provider BitYota announced an update of its flagship DWS for Big Data analytics platform, which includes data collection framework, an in-database processing pipeline for extract-load-transform (ELT), enhanced resource management and platform-specific improvements to boost analytics performance.
The company’s data collection framework provides a unified way to funnel data from a wide variety of upstream third-party application programming interface (API) sources such as Mixpanel and Flurry and NoSQL databases like MongoDB for real-time analysis.
BitYota DWS is available in multiple new configurations, including an entry-level free node with up to 1TB of storage and more powerful Premium and Enterprise offerings that can scale up from 6TB to hundreds of TBs.
“Cloud infrastructure offers an effective means to commence and scale data warehousing. Companies will consolidate data from all relevant data sources in cost effective data warehouse solutions,” Dev Patel, founder and CEO of BitYoda, told eWEEK. “Data will be collected in its native data formats and range from structured, semi-structured to unstructured. New insights will be found over raw data and not aggregated data.”
BitYota is making its MongoDB and Mixpanel extract plugins with source code available through its public Git Repository, which are available for use under the Apache 2.0 license, enabling users to modify code for their use in their environment.
In addition, numerous performance improvements enable faster loads, queries, scan and join optimizations as well as improved aggregation and exploration directly on semi-structured JavaScript Object Notation (JSON).
The company has also added the availability of compute and storage groups manageable by users, by building on BitYota’s capability to separate and elastically grow and shrink compute and storage nodes within a cluster.
Specifically, this feature collects BitYota instances running on these nodes into discrete storage or compute groups that can be assigned to individual users or business roles.
This process is designed to reduce resource contention between long and short running jobs and enables better allocation of resources to improve performance and ability to meet service-level agreements (SLAs).
The update also gives users the ability to build a custom data pipeline using SQL within the DWS that can be run on a schedule.
By using standard SQL or user-defined functions, customers can now leverage the true benefits of ELT to extract and load the data in its raw form and use the BitYota massively parallel-processing (MPP) engine for data transformations such as data quality checks, aggregations on data arrival boundaries, creation of cubes and other data manipulation tasks directly in the DWS.
On the security side, BitYota, which offers a cloud-based data warehouse service, the platform is structured on a single tenant basis, which means that each customer has their own data warehouse cluster.
“This cluster is not shared with any other customers, and the company offers user level, object level and attribute level isolation and encryption,” Patel said.