Data and Supporting Materials for "A Bounded-Memory Streaming Framework for GPU-Based Out-of-Core Reprojection of Georeferenced Raster Datasets"

This repository contains the data and supporting materials accompanying the manuscript “A Bounded-Memory Streaming Framework for GPU-Based Out-of-Core Reprojection of Georeferenced Raster Datasets.” It includes the reorganized NASA Blue Marble: Next Generation input dataset used in the experiments, the complete DOEC P1 output tile pyramid, processed and raw performance measurements, ring-cache efficiency measurements, and a minimal Visual Studio 2022 reader for the MMHI container format. The input raster was resampled to 131072×65536 pixels and stored as 2048×2048 JPEG tiles in the MMHI container. The performance measurements correspond to the experimental configurations reported in the manuscript.