MapD Technologies Adds Native Distributed and High Availability Capabilities to GPU-Powered Analytics Platform

Launch of version 3.0 enables sub-second querying and visualization of datasets with hundreds of billions of rows--and beyond

SAN FRANCISCO, April 26, 2017 /PRNewswire/  -- MapD Technologies, the leader in GPU-powered analytics, today announced the release of version 3.0 of its MapD Core database and Immerse visual analytics platform. This release includes enhanced SQL functionality, native ODBC connectivity, distributed scale-out and high availability capabilities.

MapD Core database 3.0 includes a native distributed architecture which scales data volumes and increases query performance across multiple servers. With this scale-out capability, customers can now query and visualize hundreds of billions of records with sub-second response times, even with a single rack of GPU-enabled servers. This provides significantly faster performance at lower cost than is possible even with large clusters of CPU-only analytics solutions.

"The MapD platform gives us the ability to analyze and visualize massive datasets in real-time to help provide insights into our network operations so we can improve service to Verizon customers. The distributed and high availability features of the version 3.0 release are designed to allow us to scale to even larger datasets with greater reliability," said Abdul Subhan, a lead technical architect at Verizon Wireless.

"By making it possible to scale multiple GPU-enabled servers, MapD is enabling even higher levels of performance, helping organizations drive real-time decision making," said 451 Group's senior analyst for Data Platforms & Analytics, James Curtis. "Innovations like this are making GPUs more accessible to organizations looking to take advantage of cutting-edge technologies."

The new distributed functionality is built on a shared-nothing architecture that allows near-linear performance scaling across multiple servers for complex SQL and visualization workloads. In independent SQL benchmark tests conducted by big data analyst Mark Litwintschik, MapD exhibited linear scale-out as servers were added.

"MapD was already the fastest analytics database I had ever tested, even when comparing a single server of MapD against large clusters of CPU-based solutions. With the new distributed architecture, MapD offers users record-beating performance over even more massive datasets," Litwintschik said.

The MapD 3.0 high availability capability gives customers enterprise-grade durability and redundancy by replicating data across multiple servers. Newly added records are distributed seamlessly across the cluster, while read queries are load-balanced between servers for additional throughput.

The release also includes a native ODBC driver, allowing customers to directly connect third-party tools like Tableau to the MapD Core database to take advantage of its GPU-accelerated queries for enhanced performance. This eliminates the need for an ODBC-to-JDBC bridge, as the new native ODBC connector offers higher performance and is easier to install.

MapD version 3.0 is available immediately for on-premise and cloud customers.

The release of version 3.0 follows MapD securing $25m in Series B funding from investors including New Enterprise Associates (NEA) and NVIDIA, announced earlier this month. The MapD platform is used by customers across the advertising, energy, financial services, government, retail, social media and telco sectors, including npm, Simulmedia and Verizon Wireless.

SOURCE: PR Newswire

Related Links