Hybrid Cloud and Cluster Computing for Life Sciences
Author Information
Author(s): Qiu Judy, Ekanayake Jaliya, Gunarathne Thilina, Choi Jong Youl, Bae Seung-Hee, Li Hui, Zhang Bingjing, Wu Tak-Lon, Ruan Yang, Ekanayake Saliya, Hughes Adam, Fox Geoffrey
Primary Institution: Indiana University
Hypothesis
Can a hybrid cloud and cluster computing approach improve performance for life science applications?
Conclusion
The hybrid cloud and cluster approach offers a promising environment for life sciences applications, with the Twister system providing a uniform programming model.
Supporting Evidence
- The Twister system was benchmarked against basic MapReduce and MPI in various applications.
- Initial results showed that Twister can outperform Hadoop in iterative computations.
- The study highlighted the limitations of traditional MapReduce for data mining applications.
Takeaway
This study shows that using both cloud and cluster computing together can help scientists analyze data better and faster.
Methodology
The study compared performance across Amazon, Azure, and FutureGrid using various applications developed in MPI, MapReduce, and Twister.
Limitations
The current Twister system is just a prototype, and further research is needed for scalability and fault tolerance.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website