During the days of November 28 to 30 of 2017, Big Data from Space conference was held in Toulouse, France, where more than 700 attendees gathered, including researchers, engineers, users, public institutions, infrastructure suppliers and related services people for Big Data management. This conference allowed to know the definition and evolution of Big Data, the new processes and applications of machine learning, policies of use and data security and demonstrated the strength of open data source.
Among all the topics is important to highlight the increase in data management from Copernicus EO platform, where different applications were revealed, including the large volume of data and the existing platforms for storage and pre-processing of the Sentinel 1 and 2 images, which revealed how in 2018 the volume of data will increase and will be complemented by images 1B, 2B, 3A and 5P and 3B.
They also talked about the new technologies from free open sources, such as Jupyter Notebook. This tool is recognized as one of the most important advances in the field of scientific computing, which allows to create and share documents in all programming languages. Among its variety of applications is the analysis of satellite images, machine learning, computer vision and cloud processing technologies such as Docker Container. Moreover, It should also be noted that the consolidation of new technologies represented in the following graphics increases the exploration of new platforms and the appearance of high-level architectures:
Additionally, the methodology and applications of Data Cube are highlighted, where the Europe Commissions JEODPP (JRC Earth Observation Data and Processing Platform) platform is Predominant (https://cidsecure.jrc.ec.europa.eu/home). In Order to process big data currently exists large satellite processing platforms as DEA (Digital Earth Australia) built using open sources standards (http://eos.ga.gov.au), by NASA software NEXUS (Deep Data Platform), EODC (Earth Observation Data Centre for Water Resources Monitoring; https://www.eodc.eu). And by Brazil e-Sensing: Big Earth observation data analytics for land use and land cover change information (http://www.esensing.org).
It has been interesting to understand how the increase of information technologies has been focused on data extraction, and how it has been managed to innovate in new technologies to automate data analysis, visualization and use. Applying methods such as machine learning and Deep Learning, making greater the use of satellite images to perform different analyzes and studies such as, the impact of the human populations on environments, cases of analysis of deforestation carried out by the INPE in Brazil, or the prediction of the total of electrons in the ionosphere, floods detection among different cases of study applications. It should be noted the high quality of the poster session and the presentations, giving a clear view to the current research activities, developments and Big Data initiatives of the space.
Finally, one of the lessons that should be highlighted is the availability and current use of open data, being the key to the realization of open science, developments, and rapid evolution an access to allow different sciences branches work together in synergy. Really was the best opportunity to know the best and recently big data technologies, if you are interested in this amazing topics, the next conference will be in Germany on 2019. Know the new and recently world of big data for space!
Paula 
CGIAR-CSI 
You must be logged in to post a comment.