Abstract
High-density, high-sampling rate EEG measurements generate large amounts of measurement data. When coupled with sophisticated processing methods, this presents a storage, computation and system management challenge for research groups and clinical units. Commercial cloud providers offer remote storage and on-demand compute infrastructure services that seem ideal for outsourcing the usually burst-like EEG processing workflow execution. There is little available guidance, however, on whether or when users should migrate to the cloud. The objective of this paper is to investigate the factors that determine the costs of on-premises and cloud execution of EEG workloads, and compare their total costs of ownership. An analytical cost model is developed that can be used for making informed decisions about the long-term costs of on-premises and cloud infrastructures. The model includes the cost-critical factors of the computing systems under evaluation, and expresses the effects of length of usage, system size, computational and storage capacity needs. Detailed cost models are created for on-premises clusters and cloud systems. Using these models, the costs of execution and data storage on clusters and in the cloud are investigated in detail, followed by a break-even analysis to determine when the use of an on-demand cloud infrastructure is preferable to on-premises clusters. The cost models presented in this paper help to characterise the cost-critical infrastructure and execution factors, and can support decision-makers in various scenarios. The analyses showed that cloud-based EEG data processing can reduce execution time considerably and is, in general, more economical when the computational and data storage requirements are relatively low. The cloud becomes competitive even in heavy load case scenarios if expensive, high quality, high-reliability clusters would be used locally. While the paper focuses on EEG processing, the models can be easily applied to CT, MRI, fMRI based neuroimaging workflows as well, which can provide guidance to the wider neuroimaging community for making infrastructure decisions.
Highlights
Electroencephalography (EEG) is a non-invasive, portable and cost-effective measurement technology that can provide a view with sub-millisecond resolution into the activity of the brain
While there can be found several sophisticated vendor-specific total cost of ownership (TCO) calculators for large data centre and compute facility cost calculations [56] and even for cloud infrastructures [57], the focus here is on the development of a simple yet reliable model for the neuroscience community that can be used by practicing researchers who need efficient and costeffective EEG data analysis systems
As cloud technology turns ubiquitous and equipment grants become ever more scarce, the neuroimaging community is actively investigating how cloud solutions could decrease the cost of research, while at the same time reduce execution time, increase productivity, promote data and computer program sharing across research groups
Summary
Electroencephalography (EEG) is a non-invasive, portable and cost-effective measurement technology that can provide a view with sub-millisecond resolution into the activity of the brain. The majority of EEG experiments are group studies in which individual subject data can be analysed simultaneously, independently from others Such style of parallel job execution is ideal for multi-node clusters and require no modification in the existing analysis programs. Cluster infrastructure is mainly used for very fast data storage [22] and, optionally, for computational purposes (epilepsy surgery) [23] Another purpose of using multi-processor clusters is the execution of sophisticated single subject analyses, whose exceptionally time-consuming processing can only be reduced by highly parallel algorithms [24,25,26]. Research groups can have access to large-scale campus or institutional computing facilities, local or national HPC resources (e.g. XSEDE1 and NIH Biowulf systems in the US, or EOSC3 in Europe), this is less common in clinical environments. This paper focuses on situations where the use of local IT resources incur cost
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call