What you missed in Big Data: Democratizing analytics
Accessibility has turned out to be the unifying theme for analytics this week, with a slew of updates pushing the envelope on ingesting and processing data. Taking the cake is Facebook’s decision to open-source Haxl, an internally-developed library designed to make it easier to pull information from multiple sources.
Written in Haskell, a language that is particularly well-suited for analytics thanks to its expressive syntax and high-performance compiler, the framework that serves as an abstraction between applications and the backend databases or web services they access. The toolkit makes it possible for developers to run a single query across multiple systems and cache requests for future use, a combination that Facebook says greatly simplifies the process of fetching remote information.
Like the social networking giant, embedded business intelligence (BI) provider Jaspersoft sees open-source software as a means of making Big Data more accessible for application developers. The latest version of the company’s freemium platform, which was announced in the same timeframe as Haxl and marks its first product update since joining TIBCO in April for $185 million, introduces enhanced connectors that allow users to directly access without any manual integration.
More notably, Jaspersoft 5.6 comes with a new tool called Visualize.js that lets developers embed the in-memory processing and data visualization features of the company’s open reporting server server into their applications. It’s complemented by support for new chart types and new capabilities for building custom algorithms.
Shortly after Jaspersoft announced the latest release of its flagship solution, SiSense announced that it has raised another $30 million in funding to expand into new markets and drive adoption of its namesake platform, which makes large-scale analytics more accessible to traditional enterprises. That is accomplished with a homegrown homegrown columnar database that utilizes CPU cache, RAM or disk depending on which resource type is most appropriate for the specific task at hand to improve performance while drastically reducing hardware requirements.
