Zero-Latency Rendering of Big Data Visualizations
Heavy Render works server-side to bring you interactive visualizations of high-cardinality data. It uses GPU buffer caching, modern graphics APIs, and an interface based on Vega Visualization Grammar to generate custom pointmaps, heatmaps, choropleths, scatterplots, and other visualizations, enabling zero-latency visual interaction at any scale. Render enables an immersive data exploration experience by creating and sending lightweight PNG images to the web browser, avoiding large data volume transfers. Analysts can see metadata within the visualizations as if the data were browser-side because HEAVY.AI runs imperceptibly fast SQL queries.
Server-Side Data Rendering
HEAVY.AI leverages modern graphics APIs on multiple GPUs to interactively render visualizations of data at scale. Network bandwidth is a bottleneck for the uncompressed data required by complex charts, so HEAVY.AI developed technology for in-situ rendering of on-GPU query results to accelerate the visual rendering of grain-level data. This distinguishes HEAVY.AI from other technologies that execute queries quickly but transfer the results to the client for rendering, which slows overall performance.
Large Scale Rendering of Points and Polygons
The future of geospatial analysis requires fast data rendering of complex shapes on a map. HEAVY.AI can import and display millions of lines or polygons on a geo chart with minimal lag time. This exploits the same server-side rendering technology that has always distinguished HEAVY.AI's zero-latency pointmap visualization, and it avoids any slowdowns associated with transferring high cardinality data, including large polygon data, over the network to the client.
HEAVY.AI users can select location shapes down to a local level, like census tracts or building footprints, and cross-filter interactively. Chart the heights of all buildings on a city block, or select bodies of water and graph their seasonal water temperatures.
Visualization with Vega API
Complex server-side visualizations are specified via an adaptation of the Vega Visualization Grammar, a declarative API developed by the creators of D3. HeavyImmerse, the interactive visualization interface of the HEAVY.AI platform, generates rendering specifications using Vega behind the scenes; however, developers building on HEAVY.AI can also generate custom visualizations using the same API. The result is an innovative and customizable visualization system that combines the agility of a lightweight frontend with the parallel power and rendering capabilities of a GPU engine.
Advanced Memory Management and GPU Caching
Since query results can be cached on the GPU, HEAVY.AI does not need to copy the result set before rendering it or using it as an input to a follow-on machine learning algorithm. This allows HEAVY.AI to render query results directly on the GPU, and removes the slowdowns due to network and GPU-to-CPU transfers. Three-tier caching further simplifies and accelerates server-side rendering.