Data Science Strategies for Multimessenger Astronomy

This article aims to identify and suggest data science strategies to strengthen scientific research in astronomy. The improvements in data workflow performance that can be provided by these strategies can be crucial to the multimessenger astronomy (MMA). A special focus is given to the treatment of raw data in the context of big data networks for BRICS astronomy initiatives. A preliminary design of a prototype that incorporates an MMA data cube into a data lake system is presented.

Key words
Astronomical data science; big data; data lake; data cube; MMA

Authorship

REINALDO R. ROSA

Lab for Computing and Applied Mathematics (LABAC), National Institute for Space Research (INPE), Av. dos Astronautas 1758, 12245-690 São José dos Campos, SP, BrazilNational Institute for Space Research (INPE)BrazilSão José dos Campos, SP, BrazilLab for Computing and Applied Mathematics (LABAC), National Institute for Space Research (INPE), Av. dos Astronautas 1758, 12245-690 São José dos Campos, SP, Brazil

http://orcid.org/0000-0002-2962-4322

E-mail: rrrosa.inpe@gmail.com

SCIMAGO INSTITUTIONS RANKINGS

Figures

Figures (3)

Thumbnail

Figure 1
Design of a data lake prototype for MMA with minimally structured raw data within the OLAP paradigm for multidimensional data cubes. The acronyms represent the following abbreviations: ADB for Astronomical Data Base; ADA for Agile Astronomical Data Archive; ACP for Astronomical Cyber Platforms; EM for Electromagnetic Counterparts and GW for Gravitational Waves. Raw data in this figure must be understood within the ETL paradigm reporting all dataset not yet incorporated into a data warehouse-like repository.

Thumbnail

Figure 2
Today, the data life cycle can follow the path defined by data science based entirely on cloud computing. The observer makes use of IoT devices to place their data in the cloud. The IoT is generating Big Data where your data is now inserted. A user can mine the data and extract knowledge directly in the cloud. This knowledge can serve to generate new data and thus boost the cycle.

Thumbnail

Figure 3
The graph shows the evolution of useful data generated from major astronomical observational projects. Useful data, in the context of data science, are those to be extracted automatically from the raw data workflow which, in technical language, is called Data Lake. These values are generally inferred from the estimates provided for each instrument under the condition of full operation. In the new Data Lake approach, it is reasonable to consider that up to 25% of raw data will be discarded for several reasons involving pre-processing parallelization, quality or priority.

Figure 1 Design of a data lake prototype for MMA with minimally structured raw data within the OLAP paradigm for multidimensional data cubes. The acronyms represent the following abbreviations: ADB for Astronomical Data Base; ADA for Agile Astronomical Data Archive; ACP for Astronomical Cyber Platforms; EM for Electromagnetic Counterparts and GW for Gravitational Waves. Raw data in this figure must be understood within the ETL paradigm reporting all dataset not yet incorporated into a data warehouse-like repository.

Figure 2 Today, the data life cycle can follow the path defined by data science based entirely on cloud computing. The observer makes use of IoT devices to place their data in the cloud. The IoT is generating Big Data where your data is now inserted. A user can mine the data and extract knowledge directly in the cloud. This knowledge can serve to generate new data and thus boost the cycle.

Figure 3 The graph shows the evolution of useful data generated from major astronomical observational projects. Useful data, in the context of data science, are those to be extracted automatically from the raw data workflow which, in technical language, is called Data Lake. These values are generally inferred from the estimates provided for each instrument under the condition of full operation. In the new Data Lake approach, it is reasonable to consider that up to 25% of raw data will be discarded for several reasons involving pre-processing parallelization, quality or priority.

How to cite

copy

Academia Brasileira de Ciências Rua Anfilófio de Carvalho, 29, 3º andar, 20030-060 Rio de Janeiro RJ Brasil, Tel: +55 21 3907-8100 - Rio de Janeiro - RJ - Brazil
E-mail: aabc@abc.org.br

Acompanhe os números deste periódico no seu leitor de RSS

Versão para download de PDF

PDF

Inglês

Versões e tradução automática

Versão original do texto

English

Tradução automática

Google Translator
Microsoft Translator

Como citar

RIS

BIBTEX

Outros formatos de citação e exportação:

SciELO - Scientific Electronic Library Online
Rua Dr. Diogo de Faria, 1087 – 9º andar – Vila Clementino 04037-003 São Paulo/SP - Brasil
E-mail: scielo@scielo.org

Leia a Declaração de Acesso Aberto

Métricas

Data Science Strategies for Multimessenger Astronomy

PlumX

Mensagem

[1] E-mail: rrrosa.inpe@gmail.com