Abstract
Abstract. Spatiotemporal computation implements a variety of different algorithms. When big data are involved, desktop computer or standalone application may not be able to complete the computation task due to limited memory and computing power. Now that a variety of hardware accelerators and computing platforms are available to improve the performance of geocomputation, different algorithms may have different behavior on different computing infrastructure and platforms. Some are perfect for implementation on a cluster of graphics processing units (GPUs), while GPUs may not be useful on certain kind of spatiotemporal computation. This is the same situation in utilizing a cluster of Intel's many-integrated-core (MIC) or Xeon Phi, as well as Hadoop or Spark platforms, to handle big spatiotemporal data. Furthermore, considering the energy efficiency requirement in general computation, Field Programmable Gate Array (FPGA) may be a better solution for better energy efficiency when the performance of computation could be similar or better than GPUs and MICs. It is expected that an elastic cloud computing architecture and system that integrates all of GPUs, MICs, and FPGAs could be developed and deployed to support spatiotemporal computing over heterogeneous data types and computational problems.
Highlights
Data originated from sensors aboard satellites and platforms such as airplane, UAV and mobile systems have generated high spectral, spatial, vertical and temporal resolution data
Spatiotemporal computing would have to deal with different types of data and varied algorithms
May have been equipped with a major type of hardware, such as a cluster of MICs or graphics processing units (GPUs). This means such kind of computing infrastructure may not be flexible or elastic to cope with the needs of spatiotemporal computing
Summary
Data originated from sensors aboard satellites and platforms such as airplane, UAV and mobile systems have generated high spectral, spatial, vertical and temporal resolution data. There is greater potential and challenge to extract more accurate and significant geospatial information with these high resolution data at the level that was not possible before. High spatial resolution (HSR) sensors (Greenberg et al 2009, Sridharan and Qiu 2013), such as IKONOS, QuickBird, and WorldView-2,3, can provide sub-meter pixel image products, with sufficient detail to allow the delineation of individual geographic objects such as buildings, trees, roads, and grassland (often referred to as feature extraction). Light Detection and Ranging (LiDAR) sendor can offer high vertical resolution of geometry and allows the direct collection of x, y, and z coordinates of ground objects, which makes possible automatic detection of elevated features and construction of 3 dimensional (3D) models of ground surface (Rottensteiner et al 2005). The volume, velocity, and variety of hyperspectral, HSR and 3D data, along with other socioeconomic, demographic, environmental, and social media data, pose great challenge to existing geospatial software when analyzing such
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
