Abstract
In multi-core era, much software has been developed using parallel programming technology, such as OpenMP. However, with the rise of the Internet, the data is growing dramatically. Due to the limitation of computing power and memory capacity, OpenMP applications cannot satisfy the needs of user efficiently. The flexibility and scalability of the cloud provide a new way to solve this problem. MapReduce, a parallel programming model in cloud computing, provides a solution for massive data processing. So, it needs to be considered that how to refactor the legacy OpenMP code to MapReduce model. A related refactoring approach was proposed, and a prototype tool, O2MR, was developed for the above-mentioned problem in this paper. Through experiment validation, the approach is effective, and the tool is helpful in the process of refactoring.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have