MapD, a leader in GPU-powered analytics, announced significant new feature and performance enhancements to its Core database and Immerse visual analytics platform. The new capabilities extend the company’s pioneering work in using GPUs to both query and visualize billions of records with millisecond latency. The performance characteristics of MapD’s approach are anywhere from 75 to 3,500 times faster than traditional CPU-powered databases.
This release builds upon our innovative work harnessing the power of GPUs for business intelligence, geospatial and visual analytics,” said Todd Mostak, CEO and co-founder at MapD. “Across both products, MapD Core and MapD Immerse, we were able to deliver the important features, connectors and performance our clients and prospects are seeking in a next-generation analytics solution.”
New features on MapD Core, the company’s in-memory, relational database, include:
- Even faster performance for features such as high-cardinality GROUP BY operations over multiple columns as well as the addition of and powerful new text search capabilities.
- Enhanced connectivity via a thoroughly redesigned SQL Connector to pull data from other databases, improved JDBC capabilities to power third party visualization products and enhanced Hadoop integration.
- Improvements to MapD’s innovative Iris Rendering Engine. Now clients can render multiple map layers, experience enhanced polygon and shape rendering as well as Vega-driven rendering.
New features in MapD Immerse, the company’s next-generation visual analytics engine include:
- Enhanced chart creator makes it even easier for analysts to develop new views of the data. Rather than having to choose the chart type up front, users can define their data for a chart and then switch between appropriate visualization options for the chosen data types. Further, an expanded library of chart types and the ability to set custom color dimensions provide additional options to tell stories.
- The introduction of asynchronous dashboards means that an analyst can view data as soon as it’s ready, and continue their work in the rare occurrence that a chart does not load instantly.
- The introduction of powerful new custom measures, whereby users can aggregate or specify row-level calculations using any supported SQL statement.
- The addition of custom filters to precisely control functions like the SQL WHERE statement. This allows users to look at only the most relevant records, using arbitrary, flexible SQL statements to accelerate the discovery process.
- Enhanced text search functionality. This means that analysts can carry out precise text searches using Regular Expressions, allowing complex matching rules and fuzzy matching.
- Enhanced binning capabilities, now applicable to any chart, allowing users to view data by numerical groupings.
With this release, MapD Core is now able to execute significantly more complicated queries with the same millisecond latency as before,” added MapD’s VP of Product, Ed O’Donnell. “We are also pleased with the advances we have made with Immerse, where we have refactored the interface from the ground up to take advantage of the latest in web technologies, with the main UI layer written with the React framework.”
MapD’s product can be deployed on premise or in the cloud. MapD hosts an Amazon Machine Image (AMI) on the AWS marketplace and is available on IBM Softlayer and Google Cloud deployments. Details on MapD’s hardware partners can be found on the company’s website.
Using GPUs to tackle the toughest big data problems is picking up market momentum with enterprises of all sizes,” says James Curtis, Senior Analyst, Data Platforms & Analytics at 451 Research. “The need goes beyond just being able to query and analyze large volumes of data. Companies need to be able to visualize and put information into context which can be done in real-time using the processing power of GPUs.”
MapD’s products have found significant early traction across a range of big data use cases, including log analytics, GIS, business intelligence, and social media analytics. Early adopters have included Fortune 500 companies in the telecom, retail, finance, and adtech sectors.
Sign up for the free insideBIGDATA newsletter.