GPU Rasterization for Real-Time Spatial Aggregation over Arbitrary Polygons

@article{Zacharatou2017GPURF,
  title={GPU Rasterization for Real-Time Spatial Aggregation over Arbitrary Polygons},
  author={Eleni Tzirita Zacharatou and Harish Doraiswamy and Anastasia Ailamaki and Cl{\'a}udio T. Silva and Juliana Freire},
  journal={Proc. VLDB Endow.},
  year={2017},
  volume={11},
  pages={352-365}
}
Visual exploration of spatial data relies heavily on spatial aggregation queries that slice and summarize the data over different regions. These queries comprise computationally-intensive point-in-polygon tests that associate data points to polygonal regions, challenging the responsiveness of visualization tools. This challenge is compounded by the sheer amounts of data, requiring a large number of such tests to be performed. Traditional pre-aggregation approaches are unsuitable in this setting… 
Adaptive Main-Memory Indexing for High-Performance Point-Polygon Joins
TLDR
This paper proposes an adaptive polygon index that leverages true hit fltering to avoid expensive geometric computations in most cases and introduces two geospatial join algorithms: an approximate one that guarantees a user-defined precision, and an exact one that adapts to the expected point distribution.
A GPU-friendly Geometric Data Model and Algebra for Spatial Queries
TLDR
A new model that represents spatial data as geometric objects and define an algebra consisting of GPU-friendly composable operators that operate over these objects is proposed, which shows that it is orders of magnitude faster than a CPU-based implementation and outperforms custom GPU-based approaches.
Interactive Visual Exploration of Spatio-Temporal Urban Data Sets using Urbane
TLDR
Raster Join is proposed, an approach that converts a spatial aggregation query into a set of drawing operations on a canvas and leverages the rendering pipeline of the graphics hardware (GPU) to evaluate queries on the fly at interactive speeds on commodity laptops and desktops.
GPU-Powered Spatial Database Engine for Commodity Hardware: Extended Version
TLDR
Results are reported which show that SPADE is scalable and able to handle data larger than main-memory, and its performance on a laptop is on par with that other systems that require clusters or large-memory servers.
The Case for Distance-Bounded Spatial Approximations
TLDR
This vision paper advocates for approximate spatial data processing techniques that omit exact geometric tests and provide final answers solely on the basis of (fine-grained) approximations and employs a distance-based error bound.
Efficient Filters for Geometric Intersection Computations using GPU
TLDR
A new two-step geospatial filter that first creates sketches of the geometries and uses it to detect workload and then refines the sketches by the common areas of sketches to decrease the overall computations in the refine phase is introduced.
GeoBlocks: A Query-Cache Accelerated Data Structure for Spatial Aggregation over Polygons
TLDR
GeoBlocks is introduced, a novel pre-aggregating data structure that supports spatial aggregation over arbitrary polygons and outperform on-the-fly aggregation by up to three orders of magnitude, achieving the sub-second query latencies required for interactive exploratory analytics.
Scalable Spatial GroupBy Aggregations Over Complex Polygons
TLDR
This paper proposes a highly-parallelized query processing framework to efficiently compute the spatial group-by query, which has shown significant superiority over all existing techniques.
3DPro: Querying Complex Three-Dimensional Data with Progressive Compression and Refinement
TLDR
3DPro is introduced, a system that supports efficient spatial queries for complex 3D objects through a novel Filter-Progressive-Refine paradigm that outperforms the state-of-the-art 3D data processing techniques by up to an order of magnitude.
Fast Mapping onto Census Blocks
TLDR
This paper describes two approaches labeled “simple” and “fast”, which exploits the thread, vector, and memory optimizations that are possible using a low-level language (C++) and achieves similar performance on a single server.
...
...

References

SHOWING 1-10 OF 69 REFERENCES
A GPU-based index to support interactive spatio-temporal queries over historical data
TLDR
The results show that the GPU-based index obtains interactive, sub-second response times for queries over large data sets and leads to at least two orders of magnitude speedup over spatial indexes implemented in existing open-source and commercial database systems.
Hardware acceleration for spatial selections and joins
TLDR
This paper proposes a novel approach to address the problem of complex spatial database operations using efficient rendering and searching capabilities of modern graphics hardware that does not require expensive pre-processing of the data or changes to existing storage and index structures.
Nanocubes for Real-Time Exploration of Spatiotemporal Datasets
TLDR
This work shows how to construct a data cube that fits in a modern laptop's main memory, even for billions of entries, and calls this data structure a nanocube; it can be used to generate well-known visual encodings such as heatmaps, histograms, and parallel coordinate plots.
Speeding up large-scale point-in-polygon test based spatial join on GPUs
TLDR
An end-to-end system completely on Graphics Processing Units to associate points with the polygons that they fall within by utilizing massively data parallel computing power of GPUs is designed and developed.
A Raster Approximation For Processing of Spatial Joins
TLDR
A polygon approximation for spatial join processing which is called four-colors raster signature (4CRS), which shows that the need for retrieving the representation of polygons and carrying out exact geometry tests is reduced by a factor of more than two.
imMens: Real‐time Visual Querying of Big Data
TLDR
Methods for interactive visualization of big data, following the principle that perceptual and interactive scalability should be limited by the chosen resolution of the visualized data, not the number of records are presented.
High-Performance Spatial Join Processing on GPGPUs with Applications to Large-Scale Taxi Trip Data
TLDR
The designs and implementations of GPU based filtering and refinement in spatial join processing are reported and the implications of modern hardware architectures for spatial databases and GIS are discussed.
GCMF: an efficient end-to-end spatial join system over large polygonal datasets on GPGPU platform
TLDR
GCMF, an end-to-end software system, is introduced that is able to handle spatial join (with ST_Intersect operation) over non-indexed polygonal datasets with over 3 GB file size comprising more than 600, 000 polygons on a single GPU within less than 8 sec by applying innovative filter and refinement techniques.
A sorting classification of parallel rendering
TLDR
A classification scheme is described that is based on where the sort from object coordinates to screen coordinates occurs, which it is believed is fundamental whenever both geometry processing and rasterization are performed in parallel.
Hashedcubes: Simple, Low Memory, Real-Time Visual Exploration of Big Data
TLDR
The algorithms to build and query Hashedcubes are described, and how it can drive well-known interactive visualizations such as binned scatterplots, linked histograms and heatmaps, and the typical query is answered fast enough to easily sustain a interaction.
...
...