Data Interlocking: Coupling Analytics to the Data

'Big data' analytics can be defined by the requirement for flexible, high throughput computational analysis methods applied to large, heterogeneous datasets. We propose an architectural approach to 'big data' challenges in which the movement of data is minimized, and analysis met...

Full description

Saved in:
Bibliographic Details
Published in2014 IEEE/ACM 7th International Conference on Utility and Cloud Computing pp. 696 - 701
Main Authors Kowsar, Yousef, Dashnow, Harriet, Lonie, Andrew
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2014
Subjects
Big data
Bioinformatics
cloud
Cloud computing
Computer architecture
Data analysis
data-intensive
Genomics
Software-as-a-Service
utility computing
Online AccessGet full text

Cover

More Information
Summary:'Big data' analytics can be defined by the requirement for flexible, high throughput computational analysis methods applied to large, heterogeneous datasets. We propose an architectural approach to 'big data' challenges in which the movement of data is minimized, and analysis methods are implemented on the data as portable services. We term this approach 'data interlocking'. We demonstrate the feasibility of this approach through a domain specific implementation of a data interlocking architecture, in which an on-demand computational workbench provides portable high-throughput analysis methods to large genomic datasets on cloud infrastructure.
DOI:10.1109/UCC.2014.113