IEEE Geoscience and Remote Sensing Magazine - December 2022 - 131

professional maintenance, so it is of great significance to
study the rapid processing of massive remote sensing data
by using grid computing [204].
After grid computing technology was put forward, many
countries and regions launched their own grid projects,
such as NASA's Information Power Grid, the Chinese
Ministry of Education's ChinaGrid, the United Kingdom's
Grid-Enabled Integrated Earth System model, the European
Union's DataGrid, the Asia-Pacific Grid, the European
Space Agency's Grid Processing on Demand, and other grid
computing platforms [3], [205]. As a result, many scholars
have carried out research on massive remote sensing data
processing and analysis. For example, DataGrid, funded
by the European Union, is intended to build computing
and data-intensive resource grids. Nico et al. [136] used
DataGrid to build a high-performance system for largescale
remote sensing data processing and carried out global
ozone monitoring using this system.
ChinaGrid is an important project funded by the
Ministry of Education of China, and it aims to provide
high-performance services by grid computing. Some
scholars have studied the geometric correction and image
registration of massive remote sensing data based on
ChinaGrid [229], [230]. The Romanian MedioGrid grid
system integrates basic remote sensing data processing
algorithms, such as vegetation classification, water detection,
cloud detection, image classification, and image
segmentation as well as the prevention of disasters, such
as floods and forest fires, by processing meteorological
and resource satellite remote sensing images [142]. The
Wide Area Network grid project integrates distributed
resources through the InterGrid module. This platform
was used to process Environmental Satellite and RADARSAT-1
images to analyze floods [92].
Scholars have also designed more specific remote sensing
grid computing systems for certain directions of remote
sensing data processing and analysis. For example, Shen
et al. [159] used the idea of grid computing to realize the
distributed parallel computing of remote sensing data and
designed a model composed of resource registration, an efficient
algorithm, and a distributed application and then
realized the distributed high-performance processing of
remote sensing images by resource sharing. Xue et al. [206]
designed a remote sensing grid computing framework
with four layers: presentation, model, agent, and tool. This
framework regarded the data placement task and computing
task as equally important and realized the quantitative
retrieval of remote sensing data.
The workflow can integrate distributed data and algorithms
in a grid computing environment. The grid workflow
can provide an intuitive way for users without grid
computing expertise to use grid services. The workflow
avoids the bottom implementation details of the grid and
enables users to pay attention to higher-level applications.
It can improve the system performance and ease of use
and promote the popularization and application of grid
DECEMBER 2022 IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
computing for remote sensing data processing and analysis
[3], [111], [202], (Xue et al. [246]). Xue et al. [202] designed
a workflow that combines static planning with dynamic
scheduling. Aiming at load balancing between and within
workflows, a scheduling algorithm that takes into account
both data and calculation is proposed to solve the parallelism
of remote sensing algorithms on grid platforms. Xue et al.
[203] proposed a remote sensing grid workflow architecture
based on grid computing and realized remote sensing
quantitative retrieval by constructing information service
grid nodes and a grid workflow.
The processing of remote sensing big data based on
grid computing can integrate heterogeneous computing
resources well. By optimizing and sharing these resources,
it has obvious cost advantages compared with HPC platforms,
such as supercomputers. The application of grid
computing to remote sensing big data needs to solve key
technologies, such as the computing mode, resource management,
task scheduling, data organization, and stability.
There are still some problems. Not all remote sensing
processing algorithms are suitable for parallelization on
the grid. To improve some processing algorithms, it takes
a huge amount of work to achieve a large throughput, and
grid configuration affects the performance, reliability,
and security of the computing infrastructure on a large
scale [205], [236].
REMOTE SENSING CLOUD COMPUTING PLATFORM
Grid computing and cloud computing aim to integrate
distributed heterogeneous resources and provide highperformance
services. Grid computing systems are often
oriented to specific applications and tasks, while cloud
computing platforms are more versatile, so some scholars
refer to cloud computing as Grid Computing 2.0. Cloud
computing platforms are an effective way to store, access,
and analyze datasets on very powerful servers that can act
as virtual supercomputers for users. To meet the research
and application needs around the increasing amount of
remote sensing data and the strong demand for massive
data processing capabilities, the development of remote
sensing cloud computing technology and the emergence
of platforms provide unprecedented opportunities for remote
sensing big data processing and analysis: there are
massive data resources in the cloud that do not need to be
downloaded for local processing, the cloud provides batch
and interactive big data computing services, and it provides
an API without the need to install software locally
(Fu et al. [247]).
At present, remote sensing cloud computing platforms
mainly include Amazon Web Services (AWS), Azure,
Google Earth Engine (GEE), Descartes Labs, European
Open Science Cloud, Australian Geoscience Data Cube
(AGDC), Pixel Information Expert (PIE)-Engine, EarthDataMiner,
and Aliyun. These platforms integrate massive
remote sensing data and can provide powerful computing
as well as program interfaces for spatial data and
131
IEEE Geoscience and Remote Sensing Magazine - December 2022

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - December 2022
