Paper on Cloud-Based Approach for Gridding LiDAR Data Presented at CloudCom Conference

Dec 1, 2010

OpenTopography team member Sriram Krishnan presented a paper this week at the 2nd annual IEEE International Conference on Cloud Computing Technology and Science (CloudCom 2010) in Indianapolis on LiDAR digital elevation model (DEM) generation using the cloud-based MapReduce programming model. The paper, Evaluation of MapReduce for Gridding LIDAR Data, compares a MapReduce implementation of a local gridding algorithm for DEM generation with with the C++ version currently being used in the production OpenTopography LiDAR processing system. The paper emphasizes performance (and price/performance) of the two approaches, as well as implementation complexity and scaleability.


The MapReduce programming model, introduced by Google, has become popular over the past few years as a mechanism for processing large amounts of data, using sharednothing parallelism. In this paper, we investigate the use of MapReduce technology for a local gridding algorithm for the generation of Digital Elevation Models (DEM). The local gridding algorithm utilizes the elevation information from LIDAR (Light, Detection, and Ranging) measurements contained within a circular search area to compute the elevation of each grid cell. The method is data parallel, lending itself to implementation using the MapReduce model. Here, we compare our initial C++ implementation of the gridding algorithm to a MapReduce-based implementation, and present observations on the performance (in particular, price/performance) and the implementation complexity. We also discuss the applicability of MapReduce technologies for related applications.

This research was completed by the CloudStor group at San Diego Supercomputer Center in collaboration with OpenTopography, and represents an example of ongoing LiDAR data management and processing R&D being conducted at SDSC.

UPDATED 12/05/2010: CloudCom has posted videos and slides from paper presentations at the conference. Sriram's talk can be found at this link - his presentation begins on slide 31 (time = 33:17).