Graphics Processing Units (GPUs) represent a state-of-the-art acceleration technology for general-purpose computation. GPUs are based on many-core architecture that can deliver computing performance much higher than desktop computers based on Central Processing Units (CPUs). A typical GPU device may have hundreds or thousands of processing cores that work together for massively parallel computing. Basic hardware architecture and software standards that support the use of GPUs for general-purpose computation are illustrated by focusing on Nvidia GPUs and its software framework: CUDA. Many-core GPUs can be leveraged for the acceleration of spatial problem-solving.
The use of historical maps in coordination with GIS aids scholars who are approaching a geographical study in which an historical approach is required or is interested in the geographical relationships between different historical representations of the landscape in cartographic document. Historical maps allow the comparison of spatial relationships of past phenomena and their evolution over time and permit both qualitative and quantitative diachronic analysis. In this chapter, an explanation of the use of historical maps in GIS for the study of landscape and environment is offered. After a short theoretical introduction on the meaning of the term “historical map,” the reader will find the key steps in using historic maps in a GIS, a brief overview on the challenges in interpretation of historical maps, and some example applications.
Discuss appropriate applications of the various datum transformation options
Explain the difference between NAD 27 and NAD 83 in terms of ellipsoid parameters
Outline the historical development of horizontal datums
Explain the difference in coordinate specifications for the same position when referenced to NAD 27 and NAD 83
Explain the rationale for updating NAD 27 to NAD 83
Explain why all GPS data are originally referenced to the WGS 84 datum
Identify which datum transformation options are available and unavailable in a GIS software package
Define “horizontal datum” in terms of the relationship between a coordinate system and an approximation of the Earth’s surface
Describe the limitations of a Molodenski transformation and in what circumstances a higher parameter transformation such as Helmert may be appropriate
Determine the impact of a datum transformation from NAD 27 to NAD 83 for a given location using a conversion routine maintained by the U.S. National Geodetic Survey
Explain the methodology employed by the U.S. National Geodetic Survey to transform control points from NAD 27 to NAD 83
Perform a Molodenski transformation manually
Use GIS software to perform a datum transformation
Compare and contrast the impacts of different conversion approaches, including the effect on spatial components
Create a flowchart showing the sequence of transformations on a data set (e.g., geometric and radiometric correction and mosaicking of remotely sensed data)
Prioritize a set of algorithms designed to perform transformations based on the need to maintain data integrity (e.g., converting a digital elevation model into a TIN)
Discuss the importance of planning for implementation as opposed to “winging it”
Discuss pros and cons of different implementation strategies (e.g., spiral development versus waterfall development) given the needs of a particular system
Create a budget for the resources needed to implement the system
Create a schedule for the implementation of a geospatial system based on a complete design
CP-06 - Graphics Processing Units (GPUs)
Graphics Processing Units (GPUs) represent a state-of-the-art acceleration technology for general-purpose computation. GPUs are based on many-core architecture that can deliver computing performance much higher than desktop computers based on Central Processing Units (CPUs). A typical GPU device may have hundreds or thousands of processing cores that work together for massively parallel computing. Basic hardware architecture and software standards that support the use of GPUs for general-purpose computation are illustrated by focusing on Nvidia GPUs and its software framework: CUDA. Many-core GPUs can be leveraged for the acceleration of spatial problem-solving.