ICALEPCS2013 - List of Authors (Bartsch, M.H.)

Paper

Title

Page

Exploring No-SQL Alternatives for ALMA Monitoring System

1012

T.C. Shen, A.I. Aguirre, A.J. Barrientos, M.H. Bartsch, J.P.A. Ibsen, M. Merino, L.I. Peña, R. Soto
ALMA, Joint ALMA Observatory, Santiago, Chile

The Atacama Large Millimeter /submillimeter Array (ALMA) will be a unique research instrument composed of at least 66 reconfigurable high-precision antennas, located at the Chajnantor plain in the Chilean Andes at an elevation of 5000 m. This paper describes the experience gained after several years working with the monitoring system, which has the fundamental requirement to collect and storage up to 100K variables. The original design is built on top of a cluster of relational database server and network attached storage with fiber channel interface. As the number of monitoring points increases with the number of antennas included in the array, the current monitoring system has demonstrated to be able to handle the increased data rate in the collection and storage area, but the data query interface has started to suffered serious performance degradation. A solution based on no-SQL platform was explored as an alternative of the current long-term storage system, specifically mongoDB has been chosen. Intermediate cache servers based on Redis are also introduced to allow faster online data streaming of the most recent data to data analysis application and web based charts applications

Slides WECOBA06 [0.916 MB]

Paper	Title	Page
WECOBA06	Exploring No-SQL Alternatives for ALMA Monitoring System	1012
	T.C. Shen, A.I. Aguirre, A.J. Barrientos, M.H. Bartsch, J.P.A. Ibsen, M. Merino, L.I. Peña, R. Soto ALMA, Joint ALMA Observatory, Santiago, Chile
	The Atacama Large Millimeter /submillimeter Array (ALMA) will be a unique research instrument composed of at least 66 reconfigurable high-precision antennas, located at the Chajnantor plain in the Chilean Andes at an elevation of 5000 m. This paper describes the experience gained after several years working with the monitoring system, which has the fundamental requirement to collect and storage up to 100K variables. The original design is built on top of a cluster of relational database server and network attached storage with fiber channel interface. As the number of monitoring points increases with the number of antennas included in the array, the current monitoring system has demonstrated to be able to handle the increased data rate in the collection and storage area, but the data query interface has started to suffered serious performance degradation. A solution based on no-SQL platform was explored as an alternative of the current long-term storage system, specifically mongoDB has been chosen. Intermediate cache servers based on Redis are also introduced to allow faster online data streaming of the most recent data to data analysis application and web based charts applications
	Slides WECOBA06 [0.916 MB]