Home // International Journal On Advances in Software, volume 15, numbers 1 and 2, 2022 // View article
Authors:
Martin Zinner
Wolfgang E. Nagel
Keywords: Continuous information processing; Continuous aggregation; Energy efficient computation; Real-time capability; Real-time capabilities; Data Analytics; Data processing; Stream processing; Batch processing; Business Intelligence; Ad-hoc reporting; Big Data.
Abstract:
Aggregation is a special type of association, in which objects are assembled/composed together to create a more complex object. Unfortunately, a considerable part of data aggregation during information processing in industry is still carried out in nightly batch mode, taking into account all its negative side effects. In particular, before starting aggregation, all data required for computation must be available and accessible. In contrast, our method termed Continuous Information Processing Methodology (CIPM), does not assume that the data to be aggregated is fully retrieved, the aggregation can be started as soon as the data collection is initiated. During the data collection process, partial aggregated values are calculated, such that, after the data collection phase has been completed, the final aggregated values are available for real-time ad-hoc evaluation. The existing aggregation methods apply only the usual aggregation functions such as sum(), avg(), max(), min(),count() as build in functions, on the contrary the most common aggregation functions used in various field of industry and business can be easily adapted and used within CIPM. The major benefit of our method is the elimination of the extra time necessary for batch computation, as well as reduced and spread aggregational effort over the whole collection period and tightened and straightforward computational design strategies. To conclude, the CIPM supports a paradigm shift from more or less subjectively designed individualistic conceptions in software design and development towards objectively established optimal solutions.
Pages: 94 to 110
Copyright: Copyright (c) to authors, 2022. Used with permission.
Publication date: June 30, 2022
Published in: journal
ISSN: 1942-2628