Exploiting Big Data in Engineering Adaptive Cloud Services
Our research into adaptive cloud services has shown that adaptation can require
the storage and analysis of potentially large amounts of
data [martin13book, mian13pe]
. The position I argue in my talk is that big
data has an important role to play in the engineering of adaptive software
systems, in general, and adaptive cloud services, in particular.
I use our own research to support my claim. In our research into QoS-aware
management for cloud services big data analytics are required in a number of
the component services, specifically cloud provider recommendation, workload
forecasting, performance prediction and monitoring. In each of these component
services big data is used to generate models to facilitate the necessary
decision-making. We see that the big challenge in all these cases is the need
to adapt the models when the workload and/or environment changes. We also
observe that, with the creation and management of these models, model
management will be an important addition to the adaptive middleware.
|[martin13book]||P. Martin, S. Soltani, W. Powley and M. Hassannezhad,
``QoS-Aware Cloud Application Management'',
in Advances in Parallel Computing: Clouds, Big Data and Data-Intensive Computing, February 2013.
|[mian13pe]||R. Mian, P. Martin, F. Zulkernine and J. Vazquez-Poletti,
``Towards Building Performance Models for Data-intensive Workloads in Public Clouds'',
in Proc of 4th International Conference on Performance Engineering, Prague, Czech Republic, April 2013, pp. 259--270.