This is the document website for Mimir. Mimir is a MapReduce framework over MPI for supercomputing systems.

learn more about MapReduce by reading the MapReduce paper at here and learn more about MPI at mpi-forum.org/.

Getting Started

Download, compile, and install Mimir

1. Download the code
2. Generate configuration file: autoreconf -i
3. Configure the code: ./configure –perfix=/mimir/install/directory
4. Make and install: make && make install

Feel free to send us a message at gaotao@nudt.edu.cn with your feedback.

Publications

Tao Gao, Yanfei Guo, Yanjie Wei, Bingqiang Wang, Yutong Lu, Pietro Cicotti, Pavan Balaji, and Michela Taufer. Bloomfish: A Highly Scalable Distributed K-mer Counting Framework. IEEE International Conference on Parallel and Distributed Systems (ICPADS) 2017.

Tao Gao, Yanfei Guo, Boyu Zhang, Pietro Cicotti, Yutong Lu, Pavan Balaji, and Michela Taufer. Mimir: Memory-Efficient and Scalable MapReduce for Large Supercomputing Systems. The 31st IEEE International Parallel and Distributed Processing Symposium (IPDPS) 2017.

Read our publications for the design ideas.

Contributors

• Tao Gao, National University of Defense Technology and University of Delaware
• Yanfei Guo, Argonne National Laboratory
• Boyu Zhang, University of Delaware
• Pietro Cicotti, San Diego Supercomputer Center
• Yutong Lu, National Supercomputing Center in Guangzhou and Sun Yat-sen University
• Pavan Balaji, Argonne National Laboratory
• Michela Taufer, University of Delaware

Welcome to contribute to this software.